Theory and Practice of Translation in the Age of Hypertechnology
Robert de Beaugrande
Universidade Federal de Minas Gerais
Fundação de Amparo à Pesquisa do Estado de Minas Gerais
Paper at the Second International Congress of Translators
Belo Horizonte, Brasil 23-27 July, 2001
A. Theory and practice as a dialectical cycle
1. In my newest book entitled New Foundations for a Science of Text and Discourse, I have attempted to state the conditions for a science of discourse wherein the relation between theory and practice of translation would be the most general and fundamental conception for addressing a wide range of issues, including translation. If we agree to apply our basic terms quite broadly, theory is how things get represented, whereas practice is how things get done. From a purely logical standpoint, the relations between theory and practice would be dialectical, where a dialectic consists of an interactive cycle between two sides guiding or controlling each other. When the dialectic is working smoothly, the practice is theory-driven, and the theory is practice-driven; the theory predicates and accounts for the practice; and the practice specifies and implements the theory. (Beaugrande 1997). The theory of ‘democracy’, for example, predicates, in abstract terms, ‘universal equality’ and ‘justice’ for every citizen, and accounts for such institutions as education or the law as means for achieving them. In return, the practices of such institutions specify by concrete measures what is meant by ‘equality’ and ‘justice’, e.g., through ‘standard examinations’ in schools or ‘trial by jury’ in law courts.
2. However, the dialectic may not work smoothly. In one common scenario, practice runs ahead of theory, as when people pursue their goals with no clear conceptions of the means. This scenario is probably less common than appearances suggest. Closer examination might discover implicit theories of what to try and how to do things, based on such general notions as analogy, proportion, cause-effect, part-whole, and so on; for example, people often reason from previous experience with tasks or goals partially similar to the one they are facing now. They are not pursuing mere trial and error, but a process of tuning through a series of approximations which resemble the desired goal to steadily greater degrees despite still being inadequate or inappropriate in some respects.
3. In the converse scenario, theory runs ahead of practice, as when people envision tasks and goals for which they do not command the necessary knowledge or skills. A common example in modern culture is the drive to ‘theorize’ without proper regard for the practical consequences, as is common for the ‘academic’ knowledge prized in university education but not, or no longer, relevant to professional careers.
4. In a third common scenario, theory runs away from practice. Your theory does not and cannot make progress in practice because it has either run too far ahead or has run off in some wholly impractical direction. Thus, modern education in theory prepares people to be active, knowledgeable, and self-reliant citizens, but in practice — through the process of what Paulo Freire has called educação bancária — prepares them for a passive, ignorant state of domination. In return, the process which would put the theory into genuine practice — what Freire has called educação libertadora — is accused of being ‘purely theoretical, naïve, and idealistic’ or even, as in the times of the golpe militar in Brazil, ‘dangerously anarchist’.
5. A theory that has run away from practice is often retained as the official theory, which is publicly declared but is not even intended to guide practice, whereas the operational theory is not declared or is even publicly denied but does guide practice. In ‘modern society’, our official theories are inclusive; our operational theories are exclusive; and our practices include the few and exclude the many. So ‘freedom’ coexists with domination; ‘equality’ with discrimination; ‘universal democracy’ with elitist autocracy; the ‘economic growth’ of the few with the economic shrinkage of the many; and ‘peace-keeping’ with war-mongering. People glibly profess the one and live by the other; and the failure to connect and to practice our official theories has become the endemic social disorder of the age.
6. The agenda of ecologism is to facilitate and improve a dialectical convergence between inclusive theories and inclusive practices. A perfect fit of theory with practice would be a state of utopia. Complete inclusiveness, freedom, or equality can never be attained, nor could a whole society ever definitively agree on a means guaranteed to practice them. The best a democracy and its institutions such as education can do is to promote them vigorously in everyday practices and implement effective measures for combating inequality and domination, e.g., by striving to achieve educação libertadora.
7. Ecologism pursues a hopeful utopia by enhancing the fit between the inclusive theories and the practices, encouraged rather than daunted by knowing we will never reach perfection. In exchange, ecologism repudiates the hopeless utopia of arguing that inclusive practices are naive, remote ideals far beyond the reach of ‘mere humans’ whose crucial nature is dictated by exclusive motives, such as selfishness, cruelty, and greed.
8. For its own practices, the agenda of ecologism can explore, expound, and promote both the ‘theoreticalness’ of human practices and the ‘practicality’ of human theories. This work would seek to deconstruct the manifold divisions in modern society between groups of ‘theoreticians’ versus ‘practitioners’ variously termed ‘educated’ and ‘uneducated’, or ‘experts’ and ‘laypersons’, or ‘intellectuals’ and ‘labourers’, or just ‘smart’ and ‘dumb’. Many practitioners, as Freire has described, are made to feel incompetent and intimidated about ‘theoretical’ matters, which must be left to accredited theoreticians with such forbidding titles as ‘the authorities’ or ‘the experts’.
9. Ecologism aspires to deconstruct such divisions by accrediting the empowering concept that all these groups possess different kinds of knowledge, and discrediting the disempowering concept that some groups just lack knowledge and possess ignorance. All groups are capable of educação libertadora by becoming conscious of what you know and what it means — what Freire has called conscientização. The urgency of this process is rapidly increasing amid the current waves of social change, which are so swift and radical that theory and practice may diverge irretrievably without an active agenda for convergence. And that agenda requires a co-ordinated programme of practical theories for developing new inclusive practices, especially for young people who will face a radically different world.
B. Language and discourse as theory and practice
10. If we apply our terms rather broadly, we can define a language as a theory of human knowledge and experience (what speakers or writers can say or talk about), and discourse as its practice (what they do say or talk about). A text would be defined as a communicative event that contributes to a discourse, defined in turn as configuration of mutually relevant texts. The participants include at least the text producer and the text receiver.
11. The practices are heavily theory-driven in the sense that discourse compels its participants to ‘theorise’ about what words mean, what people intend, what makes sense, and so on. Language is a ‘theory’ — or indeed a whole network of ‘theories’ — for ‘representing’ our ‘world’ and ourselves and each other in the world, and for constructing alternative states of the world or even whole alternative worlds. We understand each other insofar as our theories of the language have a parallel construction and become more finely tuned during discourse (§ 16). And due to this tuning, a language is a practice-driven theory that always remains in the process of being constituted.
12. For these reasons, discourse is the most theoretical practice humans can perform, and also the most efficient and effective in using the least effort for the most goals. In return, language is the most practical theory humans can devise, offering the resources to shape and guide almost any of our practical activities. Yet language as theory in some ways runs ahead of discourse as practice insofar as it implies some ultimate certainty and precision beyond what we can attain in any one communication. A close analysis of a text or discourse can usually uncover some uncertainty and imprecision, but these factors are the natural price for the openness of language to express an unlimited range and variety of ideas.
13. I would further submit that language is our prime example of a hopeful utopia with a limitless potential. It offers inexhaustible opportunities to improve our knowledge and understanding and to share it with an ever wider community. Even the single text is always an approximation, like ‘work in progress’ in the special sense (proposed here) of moving toward linking the inclusive theory of the language with more inclusive practices of discourse (cf. § 6, 20). The challenge, as a sensitive writer or translator should realise, is to sustain some sense of progress throughout the work of production well enough to decide when the text has ‘progressed’ far enough to be a suitable approximation.
14. Yet language is also our prime example of a ‘theory’ whose ‘theoreticalness’ is extremely well-hidden from most speakers and writers who practice it. They would regard discourse as a strictly ‘practical matter’; they would be surprised if we told them they possess a ‘theory of their language’ that makes them its ‘theoreticians’. No doubt this view arises because the theory can be practised so efficiently: many operations are automatic, below the level of conscious awareness, and require little attention to control. The native speaker need not attend to the ‘phonetic articulation’ and the ‘acoustic audition’ of language sounds; nor to the ‘grammatical formation’ of Nouns and Verbs; and so on. Such operations can run in parallel while the speaker consciously attends to strategic matters, e.g., choosing your expressions to ‘sound idiomatic’.
15. So discourse might appear to be a scenario of practice running ahead of theory (§ 2). Certainly, its practical aspects are more accessible and operational than its theoretical aspects. And the academic field of linguistics has for decades deployed theoretical oppositions such as ‘langue’ versus ‘parole’ or ‘competence’ versus ‘performance’ as pretexts for not addressing the practices of discourse as a subject worthy of academic or scientific inquiry. But the appearance changes radically once we have defined ‘language’ itself as an implicit theory (§ 10). What we need to supply now is a whole generation of explicit theories to account for the theoreticalness of discourse practices, including translation.
C. Text, intertext, interlingual text, interlingual intertext
16. If a text is a communicative event that contributes to a discourse (§ 10), then an intertext would be a communicative event involving multiple discourses. This intertextuality has not been adequately explored in conventional linguistics because it implies a disturbing openness of the text— a potentially endless field of connections and associations with other texts in the experience of discourse participants, as has recently become a fashionable conception in ‘post-modernism’ and ‘post-structuralism’. Worse yet, since every participant possesses a unique store of experiences with texts, a practical implication of intertextuality might be that any current text is experienced differently by each participant. This implication for practice is partly justified but without being disruptive, because the participants share very similar versions of the theory of the language. And participation in discourse is a practical process of tuning those theories to be at least temporarily more similar (§ 11).
17. Next, the interlingual text can be defined as a communicative event involving multiple languages. A commonplace example is code-switching, which is plainly a practice for which linguistics, which views each language as a free-standing system, has been slow to formulate a theory. Indeed, code-switching is so widespread precisely because it is so marvellously practical; you slip into whichever language you happen to have use for at the moment as in [1].
[1] Ele vai ficar puto se tu nem avisar que não vai, pois you might not be his cup of tea. Aqui não posso withdraw sem saber que tu tem uma vaga lá FOR SURE. (e-mail data)
These data show switches to pick up an English idiom ‘cup of tea’, which the Folha Webster’s edited by Professor Antônio Houaiss (himself a well-known translator) blandly circumscribes as ‘pessoa ou coisa do agrado de alguém’. ‘Withdraw’ is a technical term for university students; ‘for sure’ adds strong emphasis.
18. In the terms proposed here, a theory of code-switching would envision an interface of two or more theories, one each for each language involved, operating within a single discursive practice, as in [1]. The interface is a product of becoming multilingual, and thereby acquiring multiple perspectives upon knowledge and experience. Some facets thereof seem handier to express in one language than another, and code-switching spontaneously seeks them out.
19. A theory of translation would also be an interface of two or more theories, but operating within dual or multiple discursive practices, as in [2].
[2] Durchschnupfsicher. Super snuit-fast. Seguro, resistente. Encore plus résistant donc plus sûr. Soak-through-secure. (‘Tempo’ tissue package, made in Germany)
Whereas code-switching produces an interlingual text, translation produces an interlingual intertext, that is, a target text representing a specific source text in a different language. The term ‘target text’ is extremely apt here — something you aim at and hope to hit near the centre, but may not. In sample [2], the German text is probably an inaccurate neologism anyway, and should be ‘durchschneuzsicher’, that is, ‘secure against nose-blowing’ whereas ‘durchschnupfsicher’ would mean ‘secure against nose-running’. Presumably, the customer is to be reassured that blowing your runny nose will not also blow a hole in the tissue paper through which glutinous fluids can get onto your fingers or even get propelled out into the room and onto some bystander’s face, occasioning some resentment. The issue (though not the tissue) can be delicate, especially if the Germans blow their noses with the vigour and resolve that are so characteristic of their nation. Only the translation for the Dutch, a no less robust nation in the same rheumatic climate, roundly hits the target: ‘super-snot-resistent’ (and handily close to ‘snut-fast’, i.e. ‘snout-resistent’). The Portuguese text discretely promises ‘security’ and ‘resistance’ with no hint of why they’re needed. The French promises the same two virtues in ‘even greater’ measure than other tissues, and now in the logical causality so dear to their philosophers (‘resistant, therefore secure’, like ‘I snot, therefore I am’). The English dismally misses the target altogether, in addition to being gibberish in itself. It was obviously concocted from the German by people who only imagine they speak English — another quaint characteristic of the German nation — and even so is a miss, since a fully Teutonic equivalent would be ‘through-nose-run-safe’. As it stands, the customer could be mopping up any unappetising or corrosive substances which should not ‘soak through’ to one’s fingers. Perhaps cultural attitudes are involved; I have known Americans who considered the blowing of noses in any manner, to say nothing of German alacrity, to be a serious breach of etiquette and hence to be best performed in lavatories, where the emission of elsewhere indelicate substances is not judged unbearably indecorous.
20. The insight that translation is a utopian enterprise is due of course to José António Ortega y Gasset’s famous disquisition on ‘miseria y esplendor’. To him also we owe the paramount distinction, which I have renamed, between the hopeful utopia that counsels us to move ahead in an unlimited space for progress toward our goal, versus the hopeless utopia that counsels us to abandon the goal just because it cannot be ultimately reached (§ 7). But I propose to go much further by asserting the utopian nature of communication at large. To borrow from Roman Jakobson, we might see the very process of expressing meanings by the text producer as one instance of translation, and the process of understanding those meanings by the text receiver as another instance. Each process is like work in progress aiming at some approximation rather than some ultimate totality (§ 6, 13).
21. Unfortunately, many people implicitly subscribe to the hopeless utopia. They may realize the text could be better expressed, but they feel puzzled and unequal to the task of revising or reformulating it. Instead, audiences get confronted by grossly inadequate first approximations, for examples of which we look no further than the sayings of one George W. Bush [3-5]. Such texts urgently call for ‘intralingual translation’ from Bushspeak English into Intelligent English.
[3] Well, I think if you say you’re going to do something and don’t do it, that’s trustworthiness. (CNN online chat, 30.8.2000)
[3a] Trustworthy people actually do what they say they will.
[4] One of the common denominators I have found is that expectations rise above that which is expected. (Los Angeles, 27.9.2000)
[4a] A common factor I have noticed is expectations rising above what has been predicted. (?).
[5] Redefining the role of the United States from enablers to keep the peace to enablers to keep the peace from peacekeepers is going to be an assignment. (New York Times, 4.1.2001)
[5a] The United States should not just to keep the peace but keep it against stockpilers of nuclear arsenals. (??)
I feel confident about my ‘translation’ in [3a], but less so about [4a]; and [5a] was indirectly reconstructed from other passages of Bushspeak data.
22. Sometimes the hopeless utopia of all these instances of translation is explicitly argued:
[6] Human language was too gross a vehicle of thought—thought being incapable of absolute translation. He added, that as there can be no translation from one language into another which shall not scant the meaning somewhat, or enlarge upon it, so there is no language which can render thought without a jarring and a harshness somewhere. (Erewhon)
[7] The intranslatable words of divers languages are a proof, […] it being so obvious to observe a great store of words in one language which have not any that answer them in another. […] the names of more abstract and compounded ideas, […] when men come curiously to compare with those they are translated into, in other languages, they will find very few of them exactly to correspond in the whole extent of their significations. (An Essay Concerning Human Understanding)
In such gloomy, avowals, ‘language’ or ‘words’ are declared to pose unresolvable problems, whereas they in fact pose challenges to translators. What gets overlooked by writers like John Locke in [7] is that translation is in no way deflected, let alone defeated just because words do not ‘correspond in the whole extent of their significations’. That reservation even holds for words within the same language; if you examine authentic data in real texts, you find that the synonyms in the theory of semantics barely exist in the practices of discourse.
23. In corpus linguistics, this reservation is expressed in terms of colligations, i.e., typical combinations of grammatical selections; and of collocations, i.e., typical combinations of lexical selections. These two phenomena provide delicate links between ‘language’ and ‘text’ — or between langue and parole, competence and performance, and so on. They consist principally of regularities which are more specific than the language but more general than the text; and which are vital for making texts, including translations, sound ‘fluent’ or ‘idiomatic’ (§ 28).
24. Searching large databases with concordance programs soon leads us to conclude that virtually no two words in a language colligate and collocate in exactly the same ways. I stand ready to make this point with any pair of prospective synonyms you might care to suggest. But allow me to demonstrate with just one pair which, to my knowledge, has not been discussed in the fashion so far.
25. In my own British and American Writers Corpus (BAWC for short), totalling over 33 million words of authentic texts and installed as a set of libraries in a concordance program called WordPilot®, I find 660 occurrences of ‘tremendous’ and 976 occurrences of ‘enormous’. Only ‘tremendous’ was found collocating to the right hand with sounds, e.g., ‘applause, blast, burst, crash, cheers, chorus, clangour, din, hissing, noise, roar, shout, sobbing, thunder, tumult, uproar, voice’; ‘enormous’ never did so even once. Also, ‘tremendous’ was preferred for emotions, e.g., ‘ardour, boredom, casualness, cheerfulness, excitement, flurry, fuss, perturbation of the senses’, whereas I found only two such occurrences out of the 976 for ‘enormous’:
[8] Here I am only trying to describe the enormous emotions which cannot be described. (Heretics)
[9] Arthur, now that the first shock was over, was relaxing into an enormous sense of pleasure and contentment. (The Voyage Out)
To me these two sound unidiomatic (the first one a bit idiotic too), though they are certainly not wrong or incorrect.
26. The main right-hand collocates of ‘enormous’, in contrast, were concerned with quantities and dimensions, e.g., ‘mass, bulk, length, quantity, sum, size, volume’. Interestingly, ‘enormous’ was also found to collocate with animals, e.g. ‘crickets, butterflies, bumble bees, dog, bear, codfish, crab, frogs, boys’; and plants, e.g., ‘baobab, pine, tamarind, tulip-tree’; this time, ‘tremendous’ weighed in with just two, ‘bird’ and ‘whale’. A brief look the Adverbs may be of interest also. ‘Tremendously’ again collocated to the right with emotional states: ‘excited/ing, interested/ing, happy, in earnest’, but also with propriety: ‘dignified, neat, polite, politic, proper, sophisticated, tactful’. To the left side, the Verbs for what was done ‘tremendously’ were still more overtly emotional: ‘enjoy, admire like, love’. The Adverb ‘enormously’ predictably collocated to right with size, e.g., ‘large, long, fat, great, thick’; and with changes in size, e.g., ‘augmented, developed, extended, increased, swollen’. To the left, however, I found not just ‘gain, rise, magnify’, but also ‘enjoy, admire, amuse, value, fascinate’ — unexpectedly emotional.
27. If few true synonyms are found inside a language, the prospects look dim for finding them between two or more languages. Moreover, we should fundamentally question the theory of synonymy, to which dictionaries are firmly committed in practice, such that they might use ‘tremendous’ to define ‘enormous’, and vice versa. The constraints that give rise to patterns of collocation such I have just described are far more delicate than could be remotely supposed from the definition of the word in a dictionary. Moreover, corpus data show the meanings of individual words being circumscribed by active usage at degrees of precision which may well not be available to unaided intuition, even for fluent native speakers; yet when the data are placed before us for inspection, we may find they were prefigured in our passive knowledge. You might say somebody is ‘tremendously fat’, or ‘enormously dignified’, but a large gallery of distinguish writers of English agree to have thing the other way round.
28. These delicate constraints are undeniably relevant for translators, since they lend real substance to the qualities that have been variously called fluent, natural or idiomatic (§ 23). But to my knowledge, they have rarely figured so far in the training of translators. More generally, they have no clear place in the lesson plans of language departments, native or foreign. And they can only be extracted from browsing large banks of authentic texts, at least one bank for each language to be involved in future processes of translating.
29. So the role of technology in the theory and practice of translation is about to enter a new phase — the age of the multilingual corpus or multiconcordance. Using the MultiConcord program, for example, we can observe how the highly functional English colligation couldn’t help thinking [10] was translated to accommodate various contexts. The German goes for ‘she said to herself’ [10a]; the Spanish for ‘she could not stop tinking’ [10b]; and the French for ‘she couldn’t prevent herself from’ [10c].
[10] Alice was very nearly getting up and saying, ‘Thank you, sir, for your interesting story’, but she could not help thinking there must be more to come (Alice in Wonderland)
[10a] Alice war nahe daran, aufzustehen und zu sagen: ‘Besten Dank für deine wirklich interessante Lebensgeschichte’, aber dann sagte sie sich, daß doch noch einfach etwas kommen mußte
[10b] Alicia estaba dispuesta a levantarse y decir: ‘Gracias, señora, por su interesante historia’, pero no pudo dejar de pensar que algo más iba a decir la Tortuga
[10c] Alice fut sur le point de se lever en disant: ‘Je vous remercie, madame, de votre intéressante histoire’, mais elle ne put s'empêcher de penser qu'il devait sûrement y avoir une suite
Here beckons a brave new world of research for translation studies
D. Hypertext, hypermedia, hypertechnology
30. The term hypertext was apparently coined in the 1960s, in an obscure paper by Ted Nelson, for ‘an ongoing system of interconnecting documents’. Using terms I have proposed, it is an intertext whose intertextuality has been made explicit by producing a network of access links. More recently, the term hypermedia has been proposed to ‘expand the concept of hypertext to include other forms of digital information, e.g., graphic images, audio, video, and animation’, and to ‘present the material interactively’ ‘in response to the user’s choices’ (Sam Ebersole). Actually, this expansive concept was already present in the broad definition of ‘text’ established in semiotics or semiology, as opposed to the narrow notion of a written sample of language.
31. The term hypertechnology is not visibly established yet, and can subsume the technological systems that support hypertext and hypermedia. The concordance programs such as WordPilot for my British and American Writers Corpus and SARA for the British National Corpus are instances of hypertechnology that allows each component text, such as Walden or Jane Eyre, to be approached also as intertext and hypertext. The program enables us to refer the options of a language from text to text, freely using one context to help determine another.
32. For the theory of corpus research, I have proposed to distinguish an influential parameter between rich data versus sparse data, where richness denotes the potential of a context to determine the meaning of some term. For the trendy term ‘naff’, which is not listed in any dictionary I can locate, [11] (by a film critic) is sparser, whereas [12] (by a music critic) is richer, indicating it to mean ‘hopelessly dull or out of fashion’.
[11] Hoban won’t thank directors Wolf and Swenson for laying naff hands all over the finale of his whimsy and turning it into a homily. (Time Out)
[12] There are hundreds of ineffectual, half-hearted, derivative, dull, inexperienced, outdated and naff groups out there (New Musical Express)
Thus, corpus hypertechnology helps a student of language or a translator to offset sparseness in one context by borrowing richness from another context.
33. From tradition and commonsense, we have been saddled by two official theories: free translation and literal translation. Most writers on the subject assume they are simply incompatible and the translator must choose only one. Like other versions of formalism, some advocates of the ‘literal’ theory adopt a stance of moral rectitude from whence the ‘free’ theory can be suspected of ‘infidelities’. Yet by itself neither theory is practical, that is, can guide successful practice. At the one extreme, the ultimate free translator might be Sir John Falstaff, whose plans for (quite literal) infidelities sadly miscarry:
[13] Falstaff: Briefly, I do mean to make love to Ford’s wife; I spy entertainment in her; she discourses, she carves, she gives the leer of invitation; I can construe the action of her familiar style; and the hardest voice of her behaviour, to be English’d rightly, is ‘I am Sir John Falstaff’s’.
Pistol. He hath studied her well, and translated her will out of honesty into English.
34. At the other extreme, the ultimate literal translator might be an Internet service program called BabelFish, offered on the search engine AltaVista.com. The theory behind it is stupifying: translating proceeds word by word, or, with compound words, preferably piece by piece; any one meaning of a word or word-piece is as good as another; and the grammar will take of itself once you have the words. The effect in practice is especially striking when ‘translating’ from German to English. Sample [14] is a mysterious description of our own fair city; the mystery fades if we look at the original German text in [14a], and my English retranslation as [14b].
[14] Belo of horizons is not a routistic city, although she actually earns more attention. It is much impressing without oppressively to be worked. It has a responding architecture and everything is very generous.
[14a] Belo Horizonte ist keine touristische Stadt, obwohl sie eigentlich mehr Beachtung verdient. Sie ist sehr beeindruckend ohne bedrückend zu wirken. Sie hat eine ansprechende Architektur und alles ist sehr grosszügig.
[14] Belo Horizonte is not a tourist city, but it actually merits more notice. It is very impressive without feeling ponderous. It has attractive architecture and everything is on a grand scale.
The effect can be downright phantasmagorical, as in sample [15] from an advert for a ‘Nature Reservation Palmari’ in the far west of Brazil. Sample [15a] is the original German, where the pompous style (‘Rutschgefahr bei Matschbildung zu verringern’, ‘Möglichkeiten des lokalen Erhaltes’) plus typos (‘-gereichte’ for ‘-gerichte’, ‘wie’ for ‘wir’) multiply the opportunities for confusion; and [15b] is my own retranslation.
[15] All rooms are connected by high-put, considered timber bridges and wood bars around with rains to reduce as well as the slip hazard with gunk education. […] The breakfast consists fresh fruits from season in each case according to desire and eggs after selection. The lunch consists of imported meat places accompanies salads and season products manufactured by flour courts. The dinner consists of light courts, e.g. rice-did, as well as cold disks. Desserts are gone depending upon possibilities of the local receipt. We become unlimited quantities of dose beverages and beer are enough, as long as the supply is enough. At alcoholic beverages offer like you Cachaca liquor, from which Caipirinha one creates, both the supply is enough so long. Other beverages can by the customers bring along and without prize impact be consumed to become.
[15a] Sämtliche Zimmer sind durch hochgelegte, überdachte Holzbrücken und Holzstege miteinander verbunden um ein Nasswerden bei Regen, sowie die Rutschgefahr bei Matschbildung zu verringern. […] Das Frühstück besteht aus frischen Saisonfrüchten jeweils nach Wunsch, und Eier nach Wahl. Das Mittagessen besteht aus importierten Fleischsorten begleitet von Salaten und mit Saisonprodukten hergestellt aus Mehlgerichten. Das Abendessen besteht aus leichten Gerichten, wie z.B. Reisgereichte, sowie kalte Platten. Desserts werden je nach Möglichkeiten des lokalen Erhaltes gereicht.Wir werden unbegrenzte Mengen an Dosengetränke und Bier reichen, solange der Vorrat reicht. An alkoholischen Getränken bieten wie Ihnen Cachaca, aus welcher Caipirinha erstellt wird, beides solange der Vorrat reicht. Andere Getränke können von den Kunden mitgebracht und ohne Preisaufschlag verzehrt werden.
[15] All rooms are linked by elevated, roofed wooden bridges and ramps to prevent getting wet during rains and slipping on mud. […] Breakfast consists of a choice of fresh fruits in season, and eggs cooked as requested. Lunch consists of imported meats, salads, and seasonal rice, beans, or manioc dishes. The dinner consists of light entrees, e.g. rice-dishes, as well as cold platters. Desserts are served as locally available We will serve all the canned beverages, beer, Cachaca, and Caipirinha you want as long as the supply lasts. You are also welcome to bring your own beverages at no extra charge.
35. The results are less spectacular when Babelfish is ‘translating’ from Portuguese to English, either because the programmers worked more seriously or because the two languages are more compatible. Here are some comments by the Reitor of this university [16], and what Babelfish made of them [16a].
Of all the things my corpus data show collocating as objects of ‘fetching’ — mostly people, clothing, food, furniture, and money, plus, ‘breath, sigh, boxes, keys, doctors, police, a stoup of liquor, a cup of sack’ — nobody has ever ‘fetched excellence’, until now.
36. Babelfish is a prime example of hypertechnology run wild — an almost aleatory practice plus an absurd theory nobody seems to have seriously thought about. The theory was evidently just cobbled to accommodate lazy programming techniques: a list of individual words was stored for each language with their presumed ‘equivalents’ in the other languages, to be interchanged one by one. The designers must have expected the results to be at least intelligible, but samples like [15], and the bewilderment of the informants I have shown it to, call this expectation into question. How reliable is a ‘considered bridge?’ Is it some jungle adventure to ‘slip with gunk education’? Is it not unhealthy to be eating a dinner of ‘courts’ and ‘disks’? Should you trust hotel staff who turn into ‘beverages’ after dark?
E. Translation as alignment, approximation, and coincidence
37. Perhaps a more practical theory of translation would view it as a process of intertextual alignment. The source text as a pattern gets aligned with the target text as a pattern. The individual selections and combinations are to be performed in accord with their position and function within these patterns — therefore neither literal nor free, but positional and functional. In [15] for example, the Stretch of Text about the ‘bridges’ is positioned in the section about the accommodations, and has the function of reassuring guests who are unaccustomed to rainforest climates. This function would not be well served by either extremes of ‘literal’ [15c] or ‘free’ [15d].
[15c] Complete rooms are connected with each other by high-placed, over-roofed wood-bridges and wood-ramps in order to reduce any wet-getting during rain, and the slip-danger during mud formation.
[15d] You get around among the rooms on roofed bridges so you don’t get wet so much from the rain or slip in the mud and fall on your face so often.
The problems here reach back to the German original, where ‘verringern’ should have been ‘verhindern’. Guests don’t just want the ‘dangers reduced’ so they can get somewhat less wet in the rain and slip and fall in the mud a little less often. They want to stay dry, upright, and unmuddied, period.
38. Just as the source text is already an approximation, so too is the alignment during translation, intensified by the inherent differences between languages. But the degree of approximation can always be improved, as in the hopeful utopia described as ‘esplendor’ by Ortega y Gasset (§ 20). The translator can work through a series of approximation text in target language, each one being what Jakobson termed an intralingual translation of the previous approximation. The translator’s skill can be situated above all in the ability to manage coincidences: to exploit them when they can be found, and to generate them when they cannot. This process would be a special case of Freire’s conscientização — becoming conscious of what you know and what it means (§ 9).
39. To briefly illustrate some practices of the theory just outlined, I shall look at the two discourse types of everyday conversation and elevated poetry, which manifest very different sets of challenges. In conversation, the functions in a pattern tend to be associated with entire discourse moves in the context of situation. When Zé Carioca in the Disney comics is told the sports observer wants to talk to him [16], the encouragement of his teammate, ‘Aí, garotão!’, has, by a happy coincidence, a fairly precise functional equivalent in American English: ‘Go for it, big guy!’ When Nestor’s excuse is rejected for forgetting Gilda’s birthday [17], her ‘Essa não cola!’ — literally ‘that (feminine) doesn’t glue’ — might be translated as ‘that won’t get you off the hook!’ When Pato Donald describes the presence of a dinosaur in his house as a ‘nightmare’, the dinosaur’s reply ‘se liga!’ in [18-19] might be translated ‘back off!’
[17] ‘Zé, o olheiro da seleção quer falar com você.’ ‘Aí, garotão!’
[18] ‘Eu não esqueci,Gilda. For um error de ortografia.’ ‘Essa não cola!’
[19] ‘Que pesadelo!’ ‘Se liga! Acha que eu curto ficar preso?’
40. Obviously, cultural differences can accentuate linguistic ones. In Egyptian Arabic, a common routine starting a conversation is the greeting ‘SabaaH al-cheir’ (‘morning the-goodness’) — the ‘iDaafa’ construction of one Noun modifying another — briskly answered by ‘SabaaH al-nuur’ (‘morning the-radiance’); even more flowery greetings of the same form include ‘SabaaH al-ward’ (‘morning the-roses’) and ‘SabaaH al-full’ (‘morning the-jasmine’). For bad moods, you can use ‘SabaaH al-zift!” (morning the-pitch/tar’). In English, the culture barely allows any variation on the standard ‘Good morning’, which occurs in my BAWC database 160 times. I found just 7 occurrences of variations, none of them remotely so imaginative as ‘roses’ or ‘radiance’. I verified the contexts and found that in all cases but [24-25], the weather was indeed bright and sunny — which, to judge by the tiny quantity of data, is not common for English mornings. Also, ‘miserable morning’ [25], however accurate as meteorology, was probably an isolated and deliberate perversion.
[20] Mr. Spenlow came in, crisp and curly. ‘How are you, Copperfield?’ said he. ‘Fine morning!’ ‘Beautiful morning, sir,’ said I. (David Copperfield)
[21] An old woman came forward and stood by Stephen's elbow. ‘That’s a lovely morning, sir’, she said, ‘Glory be to God’. (Ulysses)
[22] I accosted him with what vivacity I could. ‘It is a bright, sunny morning, sir’, I said. (Jane Eyre)
[23] The chaplain of the little English church, passing at this moment, called out, ‘Fine morning, Colonel Ercott.’ (The Dark Flower)
[24] Such were the two worthies to whom Mr. Pickwick was introduced, as he took his seat at the breakfast table on Christmas morning. ‘Splendid morning, gentlemen’, said Mr. Pickwick. (Pickwick Papers)
[25] ‘How are you, old boy?’ said Tom. ‘Miserable morning’. (Pickwick Papers)
The Arabic greetings bear no relation to the weather conditions, which in Arab countries are generally radiant anyway.
41. A subtler and more interesting mode of cultural alignment can be seen in two tourist guides handed to visitors in ‘Arabia’s Wildlife Centre’ recently opened in the Emirate of Sharjah. Version [26a] is from the English guide, and version [26b] the Arabic guide, accompanied here by my own morphological translation merely for those unfamiliar with Arabic. Finally, Version [26c] is an idiomatic back-translation from the Arabic [26b].
[26a] The reptile house has exhibits of many of the Arabian snakes and lizards. […] A huge aviary, with a waterfall cascading down rocks into a small lake and river shows several species of local songbirds. […] A long corridor leads back to the entrance past enclosures containing baboons.
[26a] yabdaʔu az-Zaaʔir jawlatahu bi-iktishaafi qism az-zawaaHifi alladhi yaHtawii ʕala
he-begin the-visitor tour-his by-discovering department the-reptiles which it-contain upon
al-ʕadiid min al-afaaʕi al-ʕarabiyya was-saHaali. thumma yutabiʕu ar-riHla li-yajida nafsahu
the-variety of the-snakes the-Arabian and-the-lizards then he-continue the-journey to-he-find soul-his
Dimna qafaSin haaʔil tanHadiru fiihi shallaalaat al-maaʔ ʕala al-Sukhuur wa yaHtawi haadha
within aviary huge. they-cascade in-it falls the-water on the-rocks. and it-contains this
al-makaan al-fasiiH ʕalaa anwaaʕa mukhtalifa min al-Tuyuur al-mugharrida wa yutaabiʕu
the-place the-spacious on varieties different of the-birds the-singing. and he-continue
az-Zaaʔir riHlatahu ʕibra mamarin Tawiil yaSilu bihi ilaa Haythu tuujadu al-quruud.
the-visitor journey-his through corridor long it-take him to where exist-they the-baboons
[26c] ‘The visitor begins his tour by discovering the reptile department, which contains a variety of Arabian snakes and lizards. Then he continues the journey to find himself within a huge aviary, where waterfalls cascade on the rocks. This spacious place contains different varieties of songbirds. The visitor continues his journey through a long corridor, which takes him to where there are baboons.’
As Basil Hatim, a distinguished authority on English-Arabic translation, has pointed out, the English guide objectifies the zoo’s contents as if these existed in a situation quite independently of any visitor, who is not even mentioned. Instead, the displays and rooms are expressed as Actors of Actions like ‘having exhibits’, ‘containing species’, and ‘leading back’; the last of these is most obstinate in ignoring the visitor, who deserves to be at least the Object of the Verb. In direct contrast, the Arabic guide allots great prominence to ‘the visitor’ (‘az-Zaa’ir’) in the context of situation, who performs Actions like ‘beginning his tour’, ‘discovering the reptile department’, and ‘continuing his journey’. The cultural difference between the two guides is dramatic. In English, the objects remain static and fixed; in Arabic, the visitor is dynamic and keeps moving around.
42. The challenges of translating elevated poetry are quite different in nature, but no less imposing. In general, the functions are only marginally associated with discourse moves in the context of situation. In return, the text is constructed with a dense network of internal associations, of which rhythm and rhyme are the most prominent. Aligning two such networks is different networks is likely to test the translator’s skill to the utmost in generating coincidences. However, the theory proposed here states that such a goal is attainable in practice through successive approximations and interlingual translations, which I shall briefly demonstrate to finish off.
43. Sample [27] is the second stanza of Os Lusíadas of Luíz Vaz de Camões’, where we can easily recognise his quest for ‘um som alto e sublimado, / Um estilo grandíloquo e corrente’. [27a] was my first approximation — a ploddingly literal translation which conveys no sense of the voice and vision of Camões; sad to say, such translations of poetry have often been actually published! [2b] was my second approximation, moving toward a more elevated diction, but still without adequate attention to the networks of associations among the sounds. Finally [27c] is my third approximation — with some fourth, fifth, and sixth retouches — and may pass muster.
[27] E também as memórias gloriosas
Daqueles Reis, que foram dilatando
A Fé, o Império, e as terras viciosas
De África e de Ásia andaram devastando;
E aqueles, que por obras valerosas
Se vão da lei da morte libertando;
Cantando espalharei por toda parte,
Se a tanto me ajudar o engenho e arte.
[27a] And also the glorious memories
of those kings who went forth dilating
the faith, the empire and the vicious land
of Africa and of Asia went devastating;
and those who by valorous works
Themselves went from the law of death liberating
Singing I will spread to every part
If to so much me help the talent and art.
[27b] And I sing the mementos glorious
Of kings who voyaged hence to spread
The bounds of faith and empire,
Subjugating fierce realms of Africa and of Asia;
And those who by strength of valorous deeds
Their liberty wrest from law of death
My singing shall celebrate far and wide
With the succour of talent and art
[27c] And I sing of the memories glorious
Of the kings who sailed forth to dilate the command
Of their faith and imperial regions victorious
And of those who, by measure of valorous deed,
Their fate from the laws of mortality freed;
Subduing fell Asian and African lands;
Their fame all-embracing my singing impart
Borne afield on the wingspread of genius and art.
As for rhymes, Portuguese is at a great advantage, thanks to endings like ‘-osa’ and ‘-ando’. In English, I faced the tougher challenge of finding coincidences among rhyming stems, like ‘deed - freed’, or ‘impart - art’. As for diction, I wanted some consistency in the degree of elevation. In the original, the ‘kings’ just ‘went’; in English, I had them ‘sail forth’, since their empire was overseas, and Camões’s epic frequently invokes sea travel, e.g. in the ‘mares nunca de antes navegados’ in the very first stanza. Also, to render ‘por toda parte’ as ‘everywhere’ or ‘in every place’ wouldn’t suit; so I coined the Phrase ‘fame all-embracing’ and used the Adverb ‘afield’ to get the idea across (I first had ‘aloft’ but found it too likely to form a cliché).
44. Furthermore, I gained some elevation by putting Modifiers after Nouns, which in Portuguese is standard but in English has overtones of John Milton’s similarly elevated epic Paradise Lost, where he vows to seek an ‘adventrous Song, / That with no middle flight intends to soar / Above th’Aonian Mount, while it pursues / Things unattempted yet in Prose or Rhime’. There, we actually find the phrase ‘precious things of colour glorious’. Also, the tone was raised a bit by the old Subjunctive form ‘impart’, i.e., ‘let it impart’, as compared to the Simple Future Indicative of the original. Taken together, these networked choices might attain a consistency of style which is not merely imitated from any specific English poet, not even Milton.
45. A special challenge for me was posed by how to represent Portuguese imperialism. The locution ‘devastando […] terras viciosas’, uncomfortably seems to admit the horrendous destruction yet somehow to imply the lands deserved it. After many substitutions, I hit upon the elevated Adjective ‘fell’, which I was pleased to find the Folha Webster’s translating as ‘cruel, desumano’. As for the Verb, ‘devastating’ seems too leached out nowadays; ‘destroying’ seems too harsh; ‘exploiting’ or ‘plundering’, which would be the most accurate, don’t seem to fit what Camões had in mind; ‘subjugating’ wont fit; so I settled for ‘subduing’, which at least does not distort the actual history.
I have tried to show that theory and practice of translation need to be reconciled as one sector within a significantly larger enterprise of social and educational progress. In the past, we have all too often been hindered by operating our practices without practical theories, and dwelling too much on impractical theories. Hypertechnology promises some fresh initiatives by allowing translators to shift effortlessly among text, intertext, and hypertext, thus gaining multi-directional guidance in making their selections and combinations.
Aston G. and L. Burnard. 1998. The BNC Handbook: Exploring the British National Corpus with SARA. Edinburgh: Edinburgh University Press.
Beaugrande, R. de. 1991. Linguistic Theory: The Discourse of Fundamental Works. London: Longman.
Beaugrande, R. de. 1997. New Foundations for a Science of Text and Discourse. Stamford, CT: Ablex.
Beaugrande, R. de. 2000. Text linguistics at the millennium: Corpus data and missing links. Text 21.
Ebersole, S. 1997. Cognitive issues in the design and deployment of interactive hypermedia. Interpersonal Computing and Technology 5/1-2, 19-36.
Freire, Paulo. 1970. Pedagogia do oprimido. Rio de Janeiro: Editora Paz e Terra.
Freire, Paulo. 1980. Conscientização: Teoria e prática da libertação. São Paulo: Moraes.
King, P. and D. Woolls. 1996. Creating and using a multilingual parallel concordancer. Translation and Meaning 4:459-66.
Nelson, Ted H. 1965. Literary Machines. Self-published.
Sinclair, J.McH. 1998. “Large corpus research and foreign language teaching.” Language Policy and Language Education in Emerging Nations: Focus on Slovenia and Croatia. Ed. R. de Beaugrande, M. Grosman, and B. Seidlhofer. Stamford, CT: Ablex. Pp. 79-86.
Sinclair, J.McH. 1999. “New roles for language centres: The mayonnaise problem.” Language centres: Innovation through integration. Ed. D. Bickerton and M. Gotti. Plymouth: CercleS. Pp. 31-50
Tognini Bonelli, E. 1996. Corpus theory and practice. Pescia: Tuscan Word Centre.