Teaching Beginners To Shut Up and Listen. A Conference Interpreter Espouses Silence.
Viaggio, Sergio
Paper presented at the Annual Meeting of the First Language International Conference (Elsinore, Denmark, May 31-June 2, 1991
Abstract
All beginners at simultaneous interpreting falter at the flow of oral language, unaware that their short-term semantic flow of oral language, unaware that his short-term semantic memory can be managed more efficiently if used to store units of meaning rather than discrete words. Beginners must learn to listen for sense from the start and focus1 on the exact words used. At first, interpreters will spend too much time listening, most of the remainder processing, and then have to make do with speaking only a few words. Rather than process and speak all of the discourse, part of it incorrectly, they need to learn to say nothing. The only exception to this is when they have begun a phrase they don’t understand; at that point they should finish the phrase well, soon, and non-committally. No interpreter can translate what he or she cannot understand, but a good interpreter will manage to convey the sense without repeating each phrase. (MSE)
Teaching beginners to shut up and listen.
A conference interpreter espouses silence
by Sergio Viaggio, U.N.
The crucial insight to have come from translatoloqists is the distinction between linguistic meaning and extra-linguistic sense. The Paris school defines sense as the vouloir dire, and –French denying them a proper way of lexically distinguishing both concepts– they talk about sens and effets de sens for meaning and sense respectively. Peter Newmark has brilliantly, if –to my mind– unconvincingly, opposed this distinction. Still, if the question can be argued with respect to translation, it is definitely moot when it comes to interpretation. That I know of, with the rarest exception of two outstanding colleagues of mine who came to be interpreters down the slope of parliamentary stenography, no consecutive interpreter takes down ‘words’; an occasional key lexeme (normally only the root) yes, but that is all. What they write down is sense, and do they do it extra- linguistically! All manner of doodling miles away from any script known to anybody but themselves. By the time it is their turn to speak up, the ‘words’ they heard and understood a few seconds earlier have all but vanished. What they have in their hands is a conceptual –not semantic– framework to flesh out.
In simultaneous interpretation, the time lapse between hearing and uttering is much shorter, of course, short enough for the magic seven last words to be still stored in the short-term semantic memory.1 / It is possible, then, for the simultaneous interpreter to remember, translate and utter ‘words’. Too bad! Because it normally conspires against understanding, processing and conveying sense. The first problem the would-be interpreter faces when trying his teeth is that of managing the three competing efforts of understanding, processing and delivering2 I therefore advocate tackling them in stages from first to third. There are, in my experience, two kind & of beginners: those who have a good comprehension of oral speech and those who find it somewhat difficult to work with oral as opposed to written texts. Many beginners who have started off as translators fall within this latter category. I advise them to take the time to get used to speech and its specific variables: diction, accent, perspicuity or incompetence, etc. It shouldn’t normally take too long, provided the material is there to work with. The basic point is not to begin one’s training until one has ma e certain that oral comprehension is no problem.
But even when oral comprehension poses no difficulty, every beginner stumbles upon trying to say it all without having bothered to understand any of it. It sounds stupid, but it is actually not. What really happens is that the beginner intuitively feels that those words are well nigh instantly vanishing from his memory: if he doesn’t dispatch them right away, they’ll be lost forever! What he is not aware of is that short-term semantic memory can be managed more efficiently if instead of using it to store ‘words’, the interpreter employs it to organize chunks of sense, the unités de sens so (justifiably) dear to the Parisians.
If I ask you to remember at first hearing 49 unconnected digits, say 1-3-6-8-9-9-1-2-8-5-6-4-3-8-9-5-6-9 etc. there is absolutely no way you can do it. Now, if I mention the names of seven friends of yours, you will almost certainly be able to recall them, and if you do, you won’t have any problem in coming up with the forty-nine (seventy, if we throw the area code into the bargain) digits of their respective telephone numbers. What happens is that each series of seven (or ten) figures becomes one number, and therefore a single unit of information (interchangeable with, say, the nickname of the person) to be stored up in the short-term memory. Now let us attempt a related experiment: try and memorize on first reading the following numbers:
1774-1776- 1861-1865-1914-1918-1939-1945.
If you managed to store American Revolution-civil War-First World War-second world War you didn’t have any problem, right? What you did was reduce the 32 digits to eight numbers to four units of sense. It should have been more difficult than going from the friends’ names to their telephones.
Why? Because it took more processing. You had to make sense out of those numbers. If for whatever reason you failed to do exactly that, chances are you could not remember all of them. Notice that this kind of association is anything but linguistic: it has to do, not with our familiarity with the language (the numbers are in no specific language!), but with our knowledge of the world; not with the dictionary, but with the encyclopaedia3 We are relating and processing conceptual information which at either end of the decoding-recoding channel can and normally does acquire linguistic form. Comprehension works very much the same way. The easier it is to reduce the myriad sounds or graphic imprints to linguistic signs (the ability to understand oral or written speech), the easier it becomes to make out the linguistic meaning of the utterance (the ability to understand the language); and the sooner the words and constructions can be reduced to units of sense (the ability to understand texts), the sooner and more accurately can the hearer/reader make out the message, i.e. the sense being conveyed, at which point he won’t need the words anymore (and too bad if he does, because he simply shall not be able to remember them)• Another way of stating the same thing is to speak not in terms of sentences (linguistic) but of propositions (logical); linguistic decoding becomes therefore discourse analysis. Understanding the message rather than the words –or, if you are adamant– together with the words (but more importantly than them). Now, as we saw with our numbers above, inferring sense requires establishing a conceptual relationship between the linguistic and the extra- linguistic. The beginner must be taught to do precisely this; he must, from the very start and always ever after, be listening for sense. It is not that easy, though, in view of the many other things he thinks he’s got to do at the same time: Understanding all the words, remembering all the words; translating all the words, and saying all the words, while understanding, remembering and translating all the words that have meanwhile come in. with so many tasks in hand, who’s got spare time to understand sense! The beginner ought to be weaned from words right away. There’s no alternative.
The beginner (and would that only the beginner!) tends to cling to words, not even semantic meaning: as soon as he believes he has grasped a word he spits out the first dictionary equivalent that cores to his fretted mind. Syntax gets appallingly shortchanged and sense more often than not altogether lost or, worse, distorted. There’s only one way for him to make out sense: shut up and listen for it. There is no way to convey sense if one has not grasped it to begin wi.th. True, occasionally, word substitution can do the trick, but only at times and then awkwardly. Toboot,wordsubstitutionistoolongandcumbersome a process. There are always many short-cuts to sense; the interpreter must train himself to find them quickly and taka them without fear. since understanding gets top priority, h must do more listening than talking, and never –and I mean NEVER– allow himself to open his mouth without being reasonably sure of what he is going to say, i.e. without a plan, a strategy, which, by definition, will of necessity be conscious, the product of thought. His plan may very well consist in NOT ‘saying’ anything, just fill in what would otherwise be too long a silence with phatic language, or, better still, with information that can be safely disposed of without burning any bridges, while waiting for more ‘circumstantial evidence’ of the speaker’s communicative intentions.
The bane of the beginner is that he starts talking too soon and that he talks too much. Both go hand in hand, since if he waits longer, he won’t be able to talk that long. He must be taught –nay, forced– to listen, and to listen the right way. His constant concern should be: what is the speaker saying? why? what does he expect to achieve by it? I know it sounds like asking too much; but I am certain it is not. Unless the discourse analyst is constantly at work, the interpreter cannot hear sense. I remember my stint in Havana shortly after the 1987 earthquake in Mexico City. I had brought with me dozens of speeches taped during that session of the General Assembly; needless to say, every single one of them began with the ritual dithyrambs to the President and the secretary General, immediately followed by the condolences about the tragedy. Nine beginners –and a couple of veterans– out of tenendedupcongratulatingtheMexicansontheirearthquake. Many of them did not even realise it until they actually heard themselves over the loudspeakers. Two things had happened simultaneously: first, the interpreter had taken the speaker for granted (“Let me congratulate you, Mr. President… I should also like to express my thanks and appreciation to our secretary- General… I must as well add my condolences to the Government and people of Mexico…”) Unable to remember all the words, they retained ‘Government and people of Mexico’ and missed ‘condolences’ altogether. That should not be any problem. Back then, a few weeks after the horrible event, Mexico was synonymous with earthquake. The moment one heard ‘Mexico’, one knew –or should have known– that the earthquake was looming in the next clause. But since most of my students were not ‘thinking’, i.e. having ‘sense’ in mind; since they were not asking themselves ‘what is this man talking about and why?’ hey never paused to ponder what the Government and people of Mexico were doing in the illustrious company of the President of the General Assembly and the secretary- General. The sheer mention of Mexico should have played the trick we used to retrieve numbers: it acts with respect to the expression of condolences in a way analogous to 1939-1945 in connection to the Second World war. It activates the relevant chunk of our knowledge of the world, which, in turn, does the same thing for the interpreter’s audience: it activates their knowledge, and that is precisely the reason why the interpreter is normally able to get away with practically any ‘activating ‘ formula – exactly the same way any other speaker can4.
Some of my students tried to excuse themselves by complaining that ‘at that speed’ they did not have the time to think. They failed to realise that they did indeed: once one has understood the one word ‘Mexico’, ‘to the Government and people of‘, ‘condolences’, and ‘I would be remiss to my humanitarian duty if I failed to express my deep’ are of no avail, regardless of whether the interpreter has understood and/or remembered them; any expression of sympathy will do, the shorter the better. Naturally, this is a very special case. It happens at the very beginning of the speech, it does not require –in that situation, i.e. back then– a profound analysis, and it does not really matter how it is linguistically solved. Things become much harder when we are dealing, not with the niceties of polite society, but with the meat of an argument. Granted. But the method to approach and solve both types of instances is the same, and, as with everything else, one should learn by first applying it when it is easier.
It should be pointed out that I am not suggesting that words actually used by the speaker or the interpreter (or, indeed, the translator) never matter. Far from it. What I am saying, though, is that their relevance is secondary to the sense they are meant to make, and that, when confronted with the impossibility of rendering both words and sense, the interpreter must invariably choose the latter over the for:K· In this respect, I am as adamant as the most recalcitrant Parisians. It follows,therefore, that the beginner is to be taught to opt systematically for sense, both as a listener and as a speaker. It further follows that sense and linguistic meaning (i.e. the semantic meaning of the words as organized by the syntax of the original) seldom if ever do match one to one. There will always be more words than actually needed to convey sense, be it because the language structurally necessitates them, or because the speaker is being repetitive or expansive, either out of a legitimate rhetorical choice or through shear incompetence. It is the job of the interpreter to pursue that sense, grasp it, and then convey it. These three tasks are the non-linguistic counterparts of analysing acoustical perceptions in order to detect linguistic forms, processing the latter, and producing new linguistic material. The perception and analysis of speech should be as automatic as possible the seasoned interpreter will stop to become a are of words only when unable to ‘gloss’ over them and proceed directly from sound to sense5. Next comes the elaboration of the interpreter’s own elocution plan; he has linguistically to inform this sense (if possible and necessary- -it may be immaterial– with the closest form, semantically and stylistically). Lastly comes utterance, with proper attention to intonation, etc.
I mentioned the three competing efforts. Astonishingly enough, I have come across at least one colleague who believes it to be nonsense. A simple introspection will suffice, I hope, to prove they are there all right, and very much vying with each other, to boot. When we have trouble with the quality of the acoustic input, b e it because the speaker is looking away from the microphone, or because his accent is thick, in short, when there is ‘noise’ in the channel, we press the earphones (both of them, for once) to our temples, seal our eyelids, sit on the edge of our seats and… shut up. As soon as we decipher enough linguistic information, we send a teleqramme with it, something like: “… I … agree … with … France.” How many times this is what the audience is left with out of a speech that went “Witu regards to dah commantsu ofu pleviuspekah’s ploposu we bereave to be positivu.” How the deuce can we decode that as “With regards to the comments of the previous speaker’s proposal, we believe to be positive”, which in turn has to b e understood to mean “With regard to the proposal made by the previous speaker during his comments, we believe it is positive,” without shutting up and listening tight? What time will there be left to say “En cuanto a la propuests que el orador anterior formulara en el curao de sus comentarios, creem os que es positive?”6. Listening has used up almost all of our time and effort; processing, about seventy percent of whatever is left; elocution has to make do with three or four words. It has happened to all of us. I submit the beginner finds himself in such an extreme situation at every turn. For him, most phrases sound like the above one; he has no alternative, than, but to do as we do in those circumstances: shut up and listen, think hard and say little. And if he is left with no time to say anything at all, let him not say anything at all then: in interpreter training the alternative to the right interpretation is not a wrong one, but none whatsoever. Death by silence is better and more dignified than death by inanity. Besides, the beginner ought to know that the teacher is aware of his predicament and sympathises with him. Silence, on its part, should never mean idleness, but quiet and hard work: trying to understand, seeking to make out sense. If no sense could be made out, there is simply nothing to say.
In the case of the mention of ‘Mexico’ –as in the example of the unintelligible speaker– though faced with different problems altogether, the interpreter puts to work the same method: he looks for sense. In the first instance, ‘Mexico’ alone, regardless of its linguistic embedding, is enough to infer the preposition behind the utterance (‘the earthquake in Mexico is a very sad thing’); he does not need the rest of the words, unless he is out to come up with a formally closer translation (totally unnecessary in this specific situation, even with the Mexican delegation themselves among the audience). In the second one, no single word is enough; indeed, no single word is easily identi- fied; careful and concentrated listening allows to decipher one word here and another there, but not on the basis of phonic resemblance alone, rather, the interpreter co-relates what he hears to what he knows; phonic resemblance, as a matter of fact, enters into play ex post facto: ‘bereave’ sounds much more like ‘bereave’ than like ‘believe,’ but, since it does not make sense, the interpreter doesn’t even consider ‘bereave’ and goes on with his search7. To begin with he knows the most important thing: the speaker is not crazy; he is definitely trying to make some sense8. Grice has called this the maxim of relevance. In this, he is counting –as every normal speaker in any normal situation– on his audience being willing to understand. It is what Grice has called the maxim of ‘co-operation’. As a keenly interested listener, the interpreter is more than eager to understand; unlike any other interlocutor, he cannot simply dismiss the speaker as incompetent; he gives him the utmost benefit of the doubt. He knows that the speaker is trying to ‘say’ something, that ‘something’can be reduced to a proposition or to a hierarchical series of propositions. The semantic clues are ‘previous speaker,’ ‘proposal’ and ‘positive’. Part of the interpreter’s knowledge of the situation is that the previous speaker has been the delegate of France; therefore, what the speaker is trying to say is that he agrees with the proposal by France. That is the proposition, that –and, in the circumstances, just that– is all the interpreter needs to know and is able to say. The communication has, therefore, been assured and the interpreter has succeeded at his job. It is precisely what he is being paid for! Notice that this achievement has been possible despite the language. It would have been much easier if the speaker had merely nodded in assent: his gesture would have been much clearer than his English.
Am I advocating that interpreters never open their mouths unless they are absolutely certain that they have thoroughly understood the speaker’s sense and have completely thought out their own speech? Come on! We know better. We know when we can get away with things and when we cannot; when it is unethical to lie and when it is equally unethical not to come up with an educated guess. And we should let our students in on that. But they cannot normally allow themselves such liberty, they have to discipline themselves into listening for and making sense. Picasso did not draw square faces simply because he could not manage to draw them round. One acquires the right to bend the rules only when one has finally mastered them. There is, to my mind, only one kind of situation in which the beginner can be allowed –and even encouraged– to ‘lie’, and that is when he has started talking and hasn’t gotten the foggiest idea of where he’s going. In that case yes, the phrase must be finished, well and soon, and, most important,non-committally. Inotherwords,thebeginnerrealises that he has lied already: he has spoken as if he knew what to say and now discovers he does not; his three choices are a) to go on lying and say any monstrosity, b) to cut himself short in mid- sentence and die, or c) to stop lying but finish the sentence. It is the only time I advise my students against shutting up; when it is already too late. The only antidote is not to speak out too soon.
Mentioned has been made of Grice’s maxims of conversation (i.e. speech acts); that analysis and its development by, among others, Austin, Searle, Katz, Fodor and, more recently and relevantly to our profession, van Dijk, proves invaluable9. It reorients our search away from words and towards sense. Needless to say, good interpreters are perfectly able to do exactly that without any theoretical scaffolding, although a solid scientific base would go a long way to make them even better. The didactician, on the other hand, simply cannot do without it: in order to explain the need to listen for sense, he needs to be able himself to establish the distinction and use it. Never mind, of course, whether he has actually read Grice (I, for one, have not), or any of the others – let alone ask the students to do it; but he must be able to operate with the concepts, otherwise he won’t be in a position to instill them. There is always a reason to do or not to do things; its explanation –any explanation– is, by definition, theoretical; the didactician can, if he chooses, come up with his own insights and terminology; but what’s the point of re-inventing the wheel? Most of these things have been studied, systematized and baptized already and the literature is out there up for grabs.
At times –and much more infrequently than most interpreters believe– it is indeed necessary to ‘say it all’; what with all those Presidents and Prime Ministers and media pundits, who would dare reduce Mikhail Gorbachev’s speech to its macropropositions, right? Absolutely right! Every listener is clinging to his earphones, trying not to miss any single word or turn of phrase . But I submit that, unless the interpreter or any other listener is very much mindful precisely of the sequence and hierarchization of macropropositions and propositions, he’s bound to get lost and miss or betray sense. And I’ve got proof: At one point, Gorbachev said ‘Eto sosud bes sodershanija’, the English interpreter properly rendered it as ‘this is an empty vessel’, and I gave some vent to my poetic imagination and came up with ‘este es un continente sin contenido’. The Spanish verbatim reporters, who did not know Russian, later compared their Spanish version with the English one and noticed the discrepancy; I was duly corrected on the spot, whereupon Gorbachev ended up saying ‘esta es una embarcacion vacia’ (‘this is an empty boat’)! If you do not care about sense, words will lead you astray, ‘saying it all’ means conveying the whole of the same sense with as many of the stylistic and semantic nuances as can be possibly reproduced on the spot without abusing one’s target language. ‘Saying it a11’ presupposes, first and foremost, ‘understanding it all’, and who can ‘understand it all’ unless he has understood the gist and general drift of the speaker’s speech? Noone–andmostcertcinlynobeginner–will be ableto ‘say it all’ who cannot make out the basic propositions; whereas any good interpreter will at times find it impossible to
‘say it all’, but always manage to convey all of the sense.
‘Saying it all’ is the beginners last task, not the first one, and one should not try and teach them the other way around. Yes, I know, that is precisely how most of us were taught, but that’s no excuse for taking revenge on our students.
NOTES
1/ For the dynamics of short- and long-term memory, see Seleskovitch and Lederer (1981) and (1984).
2/ See Daniel Gile’s insightful articles.
3/ By the same token, reducing several word& to one unit of sense multiplies four- or fivefold the amount of information our short- term memory can hold for us. sense being non-linguistic, those unite can, as our friends’ numbers, be labeled ad libitum, the shorter the label the better.
4/ It is, after all, one of the basic rules of speech, and since interpretation is just a specific way of speaking, the same rules apply. I cannot refrain from quoting Mariano Garcia Landa’s gem of a definition: ‘To translate is to speak in order to say what has already been said – in another language, of course. The concept is very aptly discussed by Seleskovitch and Lederer throughout their writings.
5/ Mariano Garcia Landa has a fantastic example which I cannot resist quoting in full: “If one isolates comprehension from perception,themodeloflinguisticperceptionbecomesreducedto a sheer model of sensorial perception of linguistico-acoustic waves. To avoid such an error, we propose the theory of the perception of symbolic forms, whose essence is as follows: let the read r suppose he is driving along a highway at 80 miles an hour and, suddenly, he sees & red circle on a white background with a black number 50 in the middle. The reader slows down. A few miles down the road, a psychologist stops him and asks: what substance was that circle made of, metal, plastic or wood? what was the outer red circle’s diameter? The reader will answer that he had no time to notice those details, which, besides, are absolutely irrelevant. The only thing that matters is the symbol’s meaning [‘sense’ in Spanish!, S.V.] And yet, in order to understand the meaning of that sign –the meaning it has when standing on the right hand side of the highway, not the semantics it would have if it were lying in a municipal storehouse for traffic signs or the one it might have if it appeared in a dream– it is absolutely necessary sensorially to ‘perceive’ the physical support. The same happens with speech. ((1985a), p.181)
6/ A most ‘unfaithful’ rendering, by the way, since abhorrent Japglish has been turned into elaborate Spanish… I wonder how people who, like Peter Newmark, refuse to distinguish meaning from sense would go about translating such a speaker. I dare, moreover, my colleagues of the ‘literalist’ persuasion to reproduce that style in public!
7/ A thick foreign a cent is, precisely, the example Gile (1989) mentions as one of his triggerers of ‘deficitary’ concatenations.
8/ Peter Newmark puts it brilliantly: “…Be assured of one thing: the writer [and in our case the speaker], S.V.] must have known what he wanted to say: he would never have written a drop of nonsense in the middle of a sea of sense.” ((1988), p. 34)
9/ And also the omnipresent Garcia Landa. He cautions, however, that for him, speech acts are not limited just to •meaning meant’, but also encompass ‘meaning perceived’ ((1985a) p. 174 and (1990), chapter 2, p. 8); a crucial addition. Watch out for this latter book. I read but an incomplete, at times telegraphic torso, and believe tile, there’s a masterpiece in the making!
BIBLIOGRAPHY
BERTONE, L.: (1989) En torno de Babel. Estrategias de la interpretacion simultanea, Hachette, Buenos Aires.
DELMAS, c.: (1988) “Acces au metalinquistique natural de la lange 2 par la traduction metalinquistique”, Meta, XXXIII-3.
GARCIALANDA,M.: {1984)”Practicayteorladelainterpretacion, cuadernos de Traduccion e InterpretaciOn, 4, Barcelona.
-1984a) “Analisis del Concepto de Traduccion”, Traducao e Comunicacao, 4, Sao Paulo.
-(,985) “L’oralite de la traduction orale”, Meta, XXX-1.
DARO, v.: (1989) “The Role of Memory and Attention in Simultaneous Interpretation: An Experimental Study”, Intergreters’ Newsletter 2, Trieste.
GILE, D.: (1984) “Des difficultes de la transmission
informationelle en interpretation simultanee”, Babel, xxx-
1. -(1985) “L’interpretation de conference et la connaissance des langues: Quelquesreflexions”,Meta,XXX-4. -(1985a) “Le modele d’efforts et 1’equilibre d’interpretation en interpretation simultanee”, Meta, XXX-1. -(1988) “Le partaqe de l’attention et le ‘modele d’effort’ en interpretation simultanee”, in \Interpreters’ Newsletter, 1, Trieste.
-(1985a) La teori de la traduccion y la psicoloqia experimental de los procesos de percepcion del lenguaje, Estydios de Psicoloa1a, 19-20, Madrid.
-(1990) Teorla del bable traductoro,readfrommanuscript.
-(1988a) “An overview of Conference Interpretation Research and Theory”, proceedinga of the 29th ATA Annual conference, Learned Information, Inc., Medford, N.J.
KATZ, J. J.: (1980)PropositionalstructureandIllocutionary Force, Cambridge University Press.
LAMBERT,S.: (1988)”AHumanInformationProcessingandCognitive Approach to the Training o! Simultaneous Interpreters”, Proceedings of the 29th ATA Annual Conference, Learned Information, Inc.,Medford, N.J.
LEDERER, M.: (1976) Synecdoque et traduction, Etudes de Linguistique Appliquee, 24.
-(1985) “L’interpretation, manifestation elementaire de la traduction”, Meta, XXX-1.
MACKINTOSH, J.: ( 1985) “The Kintsch and Van Dijk Model of Discourse Comprehension and Production Applied to the Interpretation Process”, Meta, XXX-1.
NEWMARK, P.: (1988)ATextbookofTranslation,PrenticeHall, London.
RUSSO, Ch,.: (1989)”TextProcessingstrategies: AHypothesisto Assess Student’s Aptitude for Simultaneous Interpretation, Interpretation: A Neurolinguistic Approach”, Interpreters’ Newsletter, 2, Trieste.
SEARLE,J.: (1987)ExpressionandMeaning,CambridgeUniversity P ress.
–(1987) Speech Acts. An Essay in the Philosophy of Language, ibid.
SELESKOVITCH,D.: (1989)”TeachingConferenceInterpreting”,ATA Series, vol. III, SUNY, Binghampton.
SELESKOVITCH,D.andLEDERER,M.: (1981)Latraductionsimultante. Experience et theorie, Paris.
-(1984) Interpreter pour traduire, Paris, Didier Erudition. VAN DAM, I. M.: (1986) “Strategies of Simultaneous
Interpretation”, Proceedings of the 27th Annual Conference of
the ATA, Medford, NJ. VAN DIJK T. A.: (1980) Estructuras y funciones del discurso,
Siglo XXI, Mexico, 1980. -(1988) La ciencia del texto. un enfoque interdisciplinario,
Paidos, Barcelona, 1973.
VIAGGIO, S.: (1988) “Teaching Translation to Beginners, A preach continued”, Proceedings of the 29th ATA Annual Conference, Learned Information Inc., Medford, N.J.
-(1988a) “Teaching Interpretation to Beginners, or How Not to Scare then to Death”, Proceedings of the 29th ATA Annual Conference, Learned Information, Inc., Medford, N.J.
-(1989) “Teaching beginners the Blessing of Compressing and Abstracting (And Saving a Few Lives in the Process)”, Proceedings of the 30th ATA Annual Conference, Learned Information, Inc., Medford, N.J.
VIEZZI, M.: (1989) “Information Retention as Parameter for the Comparison of Sight Translation and Simultaneous Interpretation: An Experimental Study”, Interpreters’ Newsletter, 2, Trieste.