Dubbing and the Nonverbal Dimension of Translation

[Page 327] Dubbing and the nonverbal dimension of translation Patrick Zabalbeascoa, Universitat Pompeu Fabra, Barcelona 1. Introduction Translation scholars may, at different times, disagree on almost every issue of their field of interest, with, possibly, the one following exception. Both in its practice and as an object of study translation is a highly complex aspect of human activity. Indeed, we might almost say that the difficulties involved in grasping the full nature of this complexity is at the root of so much disagreement among scholars. This chapter is meant as a contribution towards showing how the translation of audiovisual texts forces us to rethink some of the recurrent issues of translation, especially a traditional division that was quite widely accepted. This division distinguished three types of translation, namely, intralingual, interlingual and intersemiotic. New forms of text production (e.g. cinema, television, hypertexts and multimedia), new discoveries and models describing and explaining language variation and semiotic factors, and even changes in social attitudes and practices, all help to show some of the limitations in the applicability of these categories. Maybe the three types mentioned above do not have such clear-cut borders as they were perceived to have. It may be more useful to think of all of the many different types of translation as a matter of degree, where many different factors come into play, so that one type of translation is distinguished from another because certain factors are more important or more restrictive in one type and not so much in another. Such an approach as this hopes to define the limits of the field of translation and translation studies; outline the parameters by which to study translations and in Fernando Poyatos (ed.) Nonverbal Communication and Translation, John Benjamins Publishing Company: Amsterdam/Philadelphia, 1997: 327-342 PATRICK ZABALBEASCOA page 328 understand the choices involved; and finally, identify the exact force of each factor for each type of translation. Thus, we can build a two-way road that will enable specific case studies to contribute towards a general theory, while general theoretical considerations can help towards a better understanding of each case. This chapter will attempt to describe some of the specific features of audiovisual translation and account for them within a model based on the variability of the priorities and restrictions involved in the translating process. 2. Linguistic approaches to translation The kind of approach expressed in Roman Jakobson’s (1959) division of the general field of translation into three parts has had a considerable influence on the way translation has been perceived ever since. According to this division, “intralingual” translation consists essentially in rewording (or paraphrasing) something within the same language. “Interlingual” translation comprises the rendering of the verbal signs of one language by means of the verbal signs of another. “Intersemiotic” translation involves the transference of a message from one kind of symbolic system to another. Typical examples of this type of translation are transmutations from a message composed of verbal signs into morse code or a flag message. Almost unanimously, the interlingual type has been regarded as ‘translation proper’, and hence studies in the field of translation concentrated on monolingual messages and how they could best be rendered into different, supposedly monodialectal, languages to make up a 100% verbal message. Most models of translation are based on this premise. Thus, the terminology of the discipline includes such terms as ‘the source language’ and ‘the target language’. The labels source-language and target-language have been widely used in the dicipline to refer to texts, cultures, readers, authors, etc. Such an approach not only seems to exclude texts that are not purely monolingual, it also implies a view of language that usually only takes verbal signs into consideration and reveals a confident feeling that there are clear-cut borders between all languages and internal stability for each one. This has lead the debate and theorization on translation to focus for a long time almost entirely on the morphological, syntactic and semantic distributions of lexical items and their differences from one language to the next. Matters to do with supra-segmental features, language variation and the combination of verbal and nonverbal elements were either eliminated by definition or swept under the carpet. Linguistics, however, gradually broadened its horizons, paying more Dubbing and the Nonverbal Dimension of Translation page 329 attention to pragmatics, for example, and developed the areas of sociolinguistics, discourse analysis and text linguistics. The results of these developments were later applied to translation studies. Translation was viewed first as something to do with the structures of languages, then with the functions of languages, and more recently, with the nature of communication, including the processes of producing and receiving texts, and, of course, their structures and functions. Discourse analysts have proved to be particularly interested in the semiotic dimension of language and authors such as Hatim and Mason (1991) have been instrumental in applying discourse models to translation. The realization that texts and utterances, rather than languages, are the objects of translation has helped tremendously in the development of the field. Furthermore, definitions of language can be made so as to account for language variation, body language and other forms of nonverbal communication. 3. Translation studies as a discipline The fact that translation can be found in so many different communication situations makes it very difficult to find a definition or a general theory that is capable of covering and accounting for all translating situations. The alternative, of course, is to give a narrow definition and eliminate from the field anything that does not fall squarely within the boundaries of the definition. Some so-called general statements about translation can only be applied to certain situations or types of translations. The only way to make them ‘general’ is to say that other types of translation are not really translation at all. Such is often the fate of dubbing and other forms of screen translation. A lot of people, both audiences and scholars, have dismissed screen translation as a ‘problem’ only of synchronization, especially of the words with the lip movement, although they admit that the problem is indeed a very big one. One consequence of viewing the matter too simplistically is that synchronization is regarded as being a unique feature of this type of translating. Another one is that any other “problems ” that there might be are not easily seen and not always taken into account. Both of these consequences have for a long time hampered the improvement of our knowledge of both screen translation and translation in general. However, synchronization is a requirement that is actually quite closely related to very traditional methods of translation, such as one-to-one equivalence and interlinear translation, which also demand a synchronization of sorts. An PATRICK ZABALBEASCOA page 330 apparent requirement for the translation of the commands of certain computer software seems to be that the translated lexical items that correspond to each command begin with the same letter as the original version, so that the same key can be pressed to activate the command in both the original and the foreign version. This sometimes leads to results that are just as bizarre as some phrases in dubbed films, which are often apparently a result of the synchronization requirement. We could even say that when translating comics, fitting foreign words into the same bubbles is a sort of spacial synchronization. Some scholars refer to the translation of certain types of texts (e.g. comics, films) as “constrained translation”. This should be used to imply that there is anything that can properly be called unconstrained translation. Dubbing is not necessarily more constrained than other forms of translation. It is rather that different forms of translation are constrained in different ways and by different factors. Relatively recently (e.g. Delabastita 1989), more systematic approaches to screen translation and other types of translation have contributed towards providing less simplistic accounts. Such approaches aim at discovering all the factors that play a role in a given translation and attempt to apply models of analysis that can help to account for as many eventualities as possible (e.g. normbased theory as developed by Toury 1995). In the case of dubbing, a large number of very different factors come into play, such as professional, sociocultural and technical ones, as well as language and communication factors, which include their nonverbal dimension. The premises used here are: (1) the basic input and output of translation are texts; (2) a film, or television production, as a text, is an audiovisual text; (3) a text is a communicative occurrence which meets seven standards of textuality (de Beaugrande & Dressler, 1981: 3), namely, cohesion, coherence, intentionality, acceptability, informativity, situationality, and intertextuality. It is also necessary to point out that there are almost as many definitions of translation as there are authors. For me, a translation is a text that fulfills three basic conditions. (1) It entails the previous existence of another text, usually called the source text (ST). (2) There is a relationship of equivalence at one or more levels between the ST and the translation (TT), i.e. the two texts are regarded as equivalent, or even the same, in some way. By virtue of the relationship of equivalence, a TT can be said to be a version of the ST. (3) There is a need and a purpose for the TT; in other words, there is a reason why a version of an ST is regarded as necessary and useful, or desirable, in a given context. One obvious need for translating a written text appears when one or more of its intended readers do not understand Dubbing and the nonverbal dimension of translation page 331 the language it is in (language is meant in its broadest sense, meaning any human form of communication, and any variety of any language). However, this is not the only possible reason. A knowledge of the language of a text is not always enough to be able to make sense of it, thus it is possible to justify the production of a more ‘accessible’ (so-called ‘intralingual’) version. Furthermore, language difference is not usually a sufficient reason, i.e. texts are not translated into foreign languages without some other justification, e.g. the ST is considered to be relevant to the TT users in some way. Equivalence, if it is going to be used as a concept in translation studies and models, cannot have an absolute, context -free value. Translational equivalence can best be defined as the result of an attempt to preserve or render a certain item or aspect of the ST in the TT. When there is a complete success in preserving a certain aspect in its entirety we may speak of ‘sameness’. Equivalence, however, is a concept that admits that we often fall short of sameness in our renderings. It is a variable in that translators are not always pursuing the same kind (or levels) of ‘equivalence’. Traditional definitions and accounts of equivalence, where it is given a constant value, lead to a prescriptive account of the product of translation. 4. The priorities and restrictions of translation Let us consider translation as a matter of priorities and restrictions, where the concept of priorities is used as a means of expressing the intended goals for a given translation task and the restrictions are the obstacles and problems that help to justify one’s choice of priorities as well as the solutions adopted in the translation. Establishing Priorities We will suppose that there is, in principle, a completely open number of potential priorities and restrictions for the whole body of existing and future translations. We will also assume that there do not necessarily have to be any universal priorities, i.e. individual priorities that are present in every translation task. Priorities and restrictions will have to be fixed anew for each task, although this may not always be as time-consuming as it sounds. A set of priorities for a given translation might be visualised on a vertical scale of importance, ranging from a very important ‘top’ priority all the way down to very minor priorities. This makes it possible to monitor the consistency with which solutions respond to higher order priorities first and foremost, and lower order priorities only in Patrick Zabalbeascoa page 332 those cases when all of the more important priorities have been satisfied first. Thus, a priority is also a restriction for all of the priorities that are ‘below’ it. So, how does one deal, for instance, with ‘stylistic equivalence’ in translation? The first thing one needs to know is the nature of the scale of priorities for the task at hand, and more precisely, what priorities are above ‘stylistic equivalence’ and which ones are below it (and, of course, what is meant by ‘stylistic equi-valence’). The position of these or any other priorities on the hierarchical scale may be different for the ST and the TT. How far can the utterance of the verbal elements of the jokes in a dubbed film deviate from perfect lip synchronization, accuracy of information given and other such considerations in pursuit of (the priority of finding) the funniest solutions? It would seem that there is often a need to strike a balance between a search for comic effect by making the translated jokes as funny as possible, on the one hand, and, on the other, finding solutions that will not put the viewer off because there is an excessive lack of synchronization; or because the plot, the structure and the coherence of the text are weakened for the sake of witty one-liners. This may also lead to the loss of other potential sources of humour such as dramatic irony. An important question is whether strict word-lip synchronization is a universal requirement or not. It has been reported (e.g. Rowe 1960) that there is a variable degree of tolerance for less-than-perfect synchronization from one audience to another. A lack of tolerance of this nature can be regarded as a restriction. At this point it is possible to deduce that the greater the tolerance the weaker the force of the need for perfect synchronization, and consequently, the wider the range of possible solutions. Restrictions, including the synchrony requirement and language differences between ST and TT, have varying degrees of force and can even be practically cancelled out at certain points of a given text. The concept of equivalence Equivalence can be seen as a means of describing the priorities for a translation. Each priority can be said to have one of the following properties: ‘equivalence’, ‘non-equivalence’, or ‘equivalence not regarded’. Equivalence is a variable and the potential levels of equivalence for a translation correspond to those priorities which ‘equivalence’ can be associated to. It is also variable in degree if we accept that its aims range from ‘absolute identity’ to ‘slight resemblance’. For example, producing comedy by translating a comedy entails that ‘intended comic effect’ is: (a) very high on the scale of importance; (b) an ‘equivalence’ priority, close to absolute identity. The word ‘intended’ as used Dubbing and the nonverbal dimension of translation page 333 above means that equivalence is a characteristic of an intention to be funny regardless of the final outcome in either the ST or the TT. What matters in this case is that the ST is seen as wanting to be funny and that is why the translator is aiming for funniness, too. Of course, this is not to say that comedy has to be translated as comedy at all times. If a translator, or his/her client wished to eliminate this aspect from the translation, for whatever reason, we might say that ‘avoid comic effect’ is a nonequivalence priority for the TT of a humorous ST. This particular possibility does not seem very likely for dubbing TV comedy into foreign languages. However, there are cases when STs of a serious nature have been turned into comic parody or farce by means of dubbing. If a priority such as ‘politically correct language and pictures’ were included as a norm or rule in all TV productions for country X, then it would have to appear in that country’s versions of foreign TV programmes, regardless of its importance in the STs. This would be an example of equivalence ‘not regarded’; even so, it will sometimes be important to know whether the result happens to produce equivalence or non-equivalence. In principle, a TT may have any number of nonequivalence priorities, but it must also reflect the equivalence priorities (variable from case to case) that can reasonably justify its existence as a version of the ST. From this angle, a TT can be judged separately on the following aspects: (a) how easily one can identify a clear set of priorities, i.e. its internal coherence; (b) how well each priority was met by the solutions provided; (c) to what extent the choice of priorities was plausible, and in tune with any other criteria it could be judged by. The dynamics of restrictions Operative restrictions are often a matter of degree (e.g. the ST — TT language ‘gap’ or the weaknesses of the ST), so we may speak of ‘the force of a restriction’. The absence of a restriction during the translating process cancels out that restriction as a factor that must be taken into consideration, which brings about a series of ‘favourable circumstances’ to be exploited by the translator. I call this phenomenon restrictions reversed. Here are seven examples of such circumstances. 1. A TT with only a small number of priorities, e.g. when translating road signs for drivers, the single most importnt priority is that they be user-friendly; and in the case of summary translation the major priorities are usually limited to finding ways of conveying the most relevant aspects the ST in a clear, brief manner. Patrick Zabalbeascoa 2. page 334 Specific characteristics of the ST, or certain aspects of it, either point very clearly to the most adequate kind of solution, or provide the translator with a greater degree of choice; e.g. STs with considerable redundancy or little ambiguity regarding meaning and intention. In the case of dubbing there are usually many nonverbal cues that help to interpret the words and the text as a whole. 3. A wide range of available priority-oriented tactics or solutions. Restrictions on availability of this sort include: lack of awareness; the presence of a series of norms that discourage or prohibit certain types of solutions; finally, solutions that depend on a certain level of technology will not be possible when the technical means are not available. In the future we may see the picture of audiovisual texts manipulated almost as much as the sound, when new technology makes that possible and convention changes. 4. Very small differences in some aspects of pairs of languages, cultural contexts, social groups, historical periods. The related restrictions are proportionally reduced. 5. The translator may be the TT initiator, the ST author or have easy access to either of them, or feel intense empathy towards the ST writer. On the other hand, the translator may have an excellent knowledge of the TT’s intended readers, what they react to, why and how, their likes and dislikes, their background knowledge or degree of expertise. 6. Adequate translator(s), i.e. the task is done by one or more people with the necessary degree of: (a) training, experience and the necessary ability to draw from it; (b) general and specialized knowledge, and, of course, an awareness of what translating involves; (c) the necessary skills to put all of this knowledge into practice for the task at hand. 7. Plenty of time, means, and incentives, and even team work, especially for dubbing. This is often a matter of money, patience and / or willpower. The nature of translation is the result of the human need to communicate, and similarities in communication strategies just as much as similar linguistic structures. The striking similarities in the structures of many languages has probably obscured this fact for a long time. In translation, we could regard differences between languages as ‘expected restrictions’, and any similarity between them as a lucky instance where the expected restriction is not operative, it is a reversed restriction. Any linguistic or cultural similarities are to be Dubbing and the nonverbal dimension of translation page 335 considered only as a possibility, but never as an inherent quality. When they do exist, we can say that certain potential restrictions have been reversed. 5. Translating for the screen This chapter does not aim to give a full account of all of the relevant factors of dubbing or subtitling. Rather, the point is to show the relevance of a few factors to do with the audiovisual nature of the sources and the products of screen translation. In a brief enumeration of the factors involved in dubbing there is: firstly, the common ground of all translating processes and general questions and aspects related to the nature of translating. Then we have to take into consideration the current dubbing techniques, means, conventions and technologies. Thirdly, the factors involved in film and television production. Fourthly, the professional context; the degree of team work (how much is required and how much there actually is). Finally, specificities of the type of film or program, e.g. in the case of comedy it is important to have a good grasp of the nature and devices of humour and, in particular, of comedy: its language, how it works, and its varieties. The translation of the words is only one part of the dubbing process. Dubbing is more different from mainstream translation in this respect than the whole question of synchronization, which can be related to the issues of literal and interlinear translation as well as syntactic and lexical equivalence. The dubbing process is something like this. A film or program is chosen and purchased. A decision is made as to whether the program should be dubbed, subtitled or not touched. If it is to be dubbed decisions have to be made regarding dubbing actors, director, studio, etc. A copy of the original version on tape is given to a freelance translator or to a number of translators, with or without the script. The translation is adjusted for timing and lip movement and performed by dubbing actors under the supervision of a dubbing director. The translator’s work often undergoes considerable changes in this stage. Finally, the new soundtrack is dubbed onto the film or tape. The words of the script constitute the verbal dimension of the sound of an audiovisual text, but there are other dimensions to the soundtrack. There are nonverbal ‘oral’ sounds including interjections and intonation patterns; other associated features of voice quality (e.g. pitch) and profile (e.g. a funny or a fearsome voice); and other sounds (special effects which can be either directly related to the words or not). The various dimensions of the soundtrack combine Patrick Zabalbeascoa page 336 in many different ways with the picture. Some images and some of the nonverbal sounds are directly related to the words uttered and others are not. Timing of delivery is an important factor in dubbing, both for lip movement and for certain situations that call for special responses with a special timing, e.g. a joke, or a dramatic climactic moment. Translating comedy for the television: priorities and restrictions A possible set of priorities for translating television situation comedy could be the following: do well in popularity ratings, be funny, aim for immediate response (i.e. be entertaining and elicit laughter), integrate the words of the translation with the other constituent parts of the audiovisual text (verbal and nonverbal units including verbal and nonverbal jokes), use language and structures (verbal and nonverbal elements) appropriate to the channel of communication. In the case of television series a specific priority is probably to create some sort of addiction or faithful following. Among the most salient restrictive factors involved in translating television programs are the following aspects: political restrictions and management policies; speed and depth of assimilation by the audience; associations and allusions between one program and another, either implicit or explicit, this being part of the required ‘shared knowledge’ of the viewers; market economy and popularity ratings; advertising, either in the form of interruptions or incorporated into the program (sponsorship). Recurrent restrictions in translating television comedy are: differences in ‘shared’ background knowledge of the two audiences, differences in moral values, cultural values, habits and traditions, differences in traditional joke-themes, the translator’s professional context, timing and lip-synchronization, synchronization of verbal and nonverbal signs, humour that depends on features of the source language, the visualization of metaphor and other similar aspects of the visual support of the text that are not allowed to be manipulated. It is true for most types of translation that there is no stability in the length or nature of the segment of text that is translated each time. Even if the length of the ‘pieces’ of text to be translated were constant they could not be interpreted independently from their context. However, it might be an interesting exercise, in order to better describe the specific nature of translating for dubbed versions, to outline the ‘units’ or chunks of text that are frequently dealt with in dubbing television comedy series. The following units could be regarded: lip movement; utterance (relevant when either the speaker is in clear view at a reasonably close distance or timing of other sounds is important); pause (the pause may be reproduced or may have to be compensated for or may have to be used to Dubbing and the nonverbal dimension of translation page 337 compensate for something else); exchange (a group of utterances, e.g. a take); a scene or sequence (as defined by the script or director of the original); chapter (e.g. a chapter of Dallas or The Black Adder); a whole series (The Black Adder) or an entire serial (Dallas). 6. Translating tactics and solutions Each part or aspect of a translation can be perceived as the outcome of a process of choosing one of various possible solutions in the light of all the operative factors of the moment. The justification for the chosen solution (and for the resulting TT as a whole) can be expressed in terms of the hierarchical structure of the priorities along with the other context -sensitive restrictions that were taken into account. Newmark (1982: 30) proposes a list of translation ‘procedures’ (i.e. tactics or types of solutions) that has three basic premises. These can be put to the test by what dubbing shows us, especially when the three are taken together. In the first place, Newmark’s procedures are based almost exclusively on lexical correspondences; secondly, the reader is given the impression that the list is closed and complete; thirdly, all translation procedures are said to be either mandatory or optional. The implication is that the degree of choice or imposition depends entirely on lexical and morphosyntactic correspondences between the language of the ST and that of the TT. A brief response to this would be that in the case of dubbing, (1) there are many nonverbal items (e.g. iconic symbols, gestures and intonation patterns) that demand attention, and therefore it would seem reasonable to look into the ‘procedures’ that have been —or could be— used in those cases; (2) a closed list of procedures such as the one Newmark proposes is certainly not a taxonomy and many solutions can belong to more than one class, so it would seem less prescriptive to consider any such list completely open to new discoveries and additions, or even a complete rearrangement; (3) the precise conditions under which a given procedure can be said to be mandatory cannot be predicted by any theoretical model, however a number of solutions or tactics can be recommended or illustrated by single-purpose stylesheets. The theory might, at most, want to suggest hypothetical solutions for consideration by translators or others involved in translation assignments in some way or other. It might be true to say that, in translation, the end justifies the means, so it is unfortunate when the means are not a clear reflection of the end that is Patrick Zabalbeascoa page 338 pursued. Depending on the specific characteristics of the translation assignment, a translator may resort to a variable range of priority-oriented tactics, or solutions. We could speak firstly of ‘textual’ tactics as making use of either ‘discursal’ or ‘paradiscursal’ solutions. Discursal solutions are those items of the TT that are inserted in the main body and discourse of the text. ‘Paradiscursal’ solutions are part of the text (especially the physical dimension of the text) but somehow independent from the main body and discourse of the text; e.g. footnotes, glossaries, appendices, forewords, illustrations. Secondly, ‘paratextual’ tactics would, for instance, involve the use by the translator of notes or directions on how the words are to be delivered by the dubbing actor for the words to be interpreted as intended by the translator; these notes would not be meant to be within the reach of the TT user. On a different level, solutions can fall into one of the following three categories: (1) purely verbal, e.g. accounting for ST words or word groups by adding, substituting or deleting words in the TT; by paraphrasing, or defining; by increasing verbal redundancy, etc.; (2) purely non-verbal: e.g. adding, changing or deleting nonverbal signs in the form of pictures, colours, gestures, mimicry, sounds, etc.; (3) verbal - nonverbal compensation, whereby an ST word or word group along with its textual value is accounted for entirely or partly in the TT by one or more nonverbal elements, or vice versa, e.g. the potential meaningfulness of certain kinds of voices or intonations. 7. The semiotic and nonverbal dimension of translating One thing that dubbing makes us more aware of is that no text can be made entirely of verbal signs because these signs always need some sort of physical support, and, as there is usually a choice as to which physical support a text is going to have and what vehicle is going to be used to transport a text from its sender(s) to its user(s), that is, its mode of discourse and channel of communica-tion. Both the physical support and the means of communication are potentially meaningful and can act as nonverbal signs. Thus, we often feel the need to interpret, for example, exactly why some letters on a page are larger than others, or why some articles in a newspaper are illustrated and others are not; or why different people or groups use different intonation patterns or gestures. Body language, facial expressions, hesitations and eye contact are factors that interpreters have to take into account while trying to render a speaker’s intended message. They are also factors that have a role to play in film translation. What does one do, for example, when the words in a novel one is translating were Dubbing and the nonverbal dimension of translation page 339 deliberately chosen by the ST author so that none of them are words that contain the letter e? It has already been pointed out that it all depends on what one wants to do, but in any case part of the answer lies in finding out why this was done in the ST and what meaning(s) this feature of the novel is supposed to carry. In a Chinese poem, a word might be chosen not only because of the beauty of its connotations and associatio ns, and not only because it has a beautiful sound when pronounced in the context of its poem, but also because it has a certain artistic quality in its visual impact on the page. One might say that this is an example of an impossible word to translate into an Indo-European language, or one might try to look for some compensatory tactic to solve the problem by means of rhyme or rhythm, for example. When interpreting a text and the relations between the verbal and nonverbal signs it is important to know whether one is subordinated to the other or whether they are meant to complement each other. For example, in translating a song, which is the result of Neruda’s poems put to music, the translator might feel that the music is subordinated to the words, or the original form of the poem. Songs that appear in films are often subtitled even when the rest of the film is dubbed, following the blanket principle that the music and the voices are always more important than the lyrics. If this were done in the example above, the TT audience would be presented with the accompanying music untouched and the words translated according to the strict restrictions of subtitling conventions. An alternative tactic that would serve a different purpose might be to use a previously published prestigious translation of the original poem that the TT audience might even recognise as a foreign version of Neruda and then put accompanying music to it and have it sung by a dubbing actor. In many cartoons, e.g. some Tom and Jerry battles, the pace of the action, possibly even the nature of the action, is conditioned by the rhythm of well known classical music, which, in turn, takes on a new meaning since its interpretation is conditioned by the image of the cartoon. What we are beginning to see is that even if we restrict translation to a purely verbal operation, nonverbal factors and their potential relevance have to be taken into account as well. The difference between an audiovisual text such as a film and other types of texts such as telegrams or novels is the relative importance of the verbal and nonverbal signs, the relationship between the two types of signs, and the amount of ‘space’ or time taken up by each type in the total ‘volume’ of a text. Dubbing reminds us of the possibility that all translations have a semiotic dimension, in some it is more important than in others. Here are two alternative Patrick Zabalbeascoa page 340 classifications to help locate dubbing and subtitling on the ‘map’ of translation activities and specializations. The first is a classification of texts according to mode of perception, including the verbal nonverbal distinction. 1. Read only, where nonverbal elements have very little relevance; e.g. a novel. 2. Read and seen (verbal and non-verbal visual signs); e.g. a comic book, or an entirely verbal text where layout, formatting and/or colours are highly relevant or meaningful (a frequent case in advertising). 3. Heard only (verbal and non-verbal sounds); e.g. a radio program. 4. Heard and seen (including verbal and non-verbal signs); e.g. a play performed on stage. 5. Heard and seen and read; e.g. a film with subtitles or with written messages in the original picture. 6. Seen and/or heard only (including only nonverbal images and sounds); e.g. a comic strip with no words, some silent films . The second classification, based on the first, is one of many possible arrangements of texts that are TTs or STs. A. Written texts to be read silently by TT readership, including summary translations. This type includes the translation of text -types 1 and 2 according to the classification above. B. Written texts to be perceived aurally, drawn from the non-spontaneous texts of text -type 3. C. Written texts to be perceived ‘audiovisually’ (the scripts for text -types 4 and 5). In subtitling, the ST may belong to text -type 4, but the TT always belongs to text -type 5. D. Spontaneous speech (not written beforehand): text -type 3 in the case of simultaneous interpreting, or text -type 4 in the case of liaison interpreting, where body language is a dimension of the communication act. Another interesting distinction can be drawn between processes where the translator is, in principle and by convention, free to manipulate almost any feature or component of the ST, as in many of the texts belonging to text -types 1 and 3; and, on the other hand, those texts where the translator can influence or manipulate only a part of what might be considered the entirety of a text, a typical restriction for audiovisual translators. Convention has it that dubbing and Dubbing and the nonverbal dimension of translation page 341 subtitling may not alter the picture of a film or freeze a frame for a split second; audiovisual texts for translation can only be manipulated or ‘translated’ in their verbal components. There are two further considerations to take into account. The first is language variation within a given text. It is true that some texts include only one variety of a single language, but there are many texts, especially, in literature, in films and on television, where many different varieties of a language (idiolects, dialects, sociolects, etc.) are included. There are also texts that have verbal elements pertaining to more than one language. The second consideration is the semiotic value of some of the verbal elements themselves, as shown in Hatim and Mason (1990), for example. 8. Conclusion From what has been presented in this chapter, I can now propose that the intralingual-interlingual distinction be superseded by the notion that both STs and TTs can manifest variable degrees of ‘monolinguality’ and membership to a single genre and a single type of discourse. Thus, interlingual translation (seen as a process whereby a monolingual ST which is prototypical of a given type of discourse and genre is rendered into a TT with same characteristics, except for a change of language) is only one type of translation, or rather it is on one end of a cline ranging from monolingual and monodiscursive prototypicality to hybrid multilingual or multidialectal texts. Instead of interlingual—intersemiotic distinction it seems more accurate to regard texts as having varying proportions of linguistic and verbal elements and nonlinguistic or nonverbal signs. In this light, all translations are the result of semiotic processes, where nonlinguistic (and/or nonverbal / suprasegmental / etc.) signs are more important in some translations than in others. Future attempts to give accounts of translation ‘techniques’ or solutions might wish to include nonverbal compensatory tactics which to date have not been paid sufficient attention and hence have not been satisfactorily described. Classes or accounts of solutions would also have to be accompanied by explanations as to whether they are hypothetical and illustrative of ‘theoretical’ possibilities, or whether they are mandatory (or optional or forbidden) for certain assignments, for example, as they would appear in a single-purpose in-house stylebook. Patrick Zabalbeascoa page 342 References and Bibliography Beaugrande, R. de; W. Dressler. 1981. Introduction to Text Linguistics. London: Longman. Caillé, P. 1960. “Cinéma et Traduction: Le Traducteur devant l’Ecran”. Babel 6:3. 103-109. Cary, E. 1960. “La Traduction Totale”, Babel 6:3, 110-115. Delabastita, D. 1989. “Translation and Mass-communication: Film and T.V. Translation as Evidence of Cultural Dynamics”, Babel 35:4. 193-218. Delabastita, D. 1989. “Translation and Mass-communication: Film and T.V. Translation as Evidence of Cultural Dynamics”. Babel 35:4; FIT 193-218. Fodor, I. 1976. Film Dubbing: Phonetic, Semiotic, Esthetic and Psychological Aspects. Hamburg: Buske. Goris, O. 1993. “The Question of French Dubbing: Towards a Frame for Systematic Investigation”, Target 5:2, 169-190, Amsterdam: John Benjamins. Halliday, M. 1985. Spoken and Written Language, Victoria, Australia: Deakin University Production Unit. Hatim, B. and I. Mason. 1990. Discourse and the Translator, Harlow: Longman. Izard, N. 1992. La Traducció Cinematogràfica, Barcelona: Generalitat de Catalunya. Jakobson, R. 1959. “On Linguistic Aspects of Translation.”, In Brower, (ed.), On Translation, Cambridge: Harvard University Press. Mayoral, R. 1994. “La Traducción cinematográfica: el subtitulado”, Sendebar, 4, 4568. Palmer, J. 1987. The Logic of the Absurd, London: BFI books. Rowe, T. 1960. “The English Dubbing Text ” Babel 6:3, 116-120. Toury, G. 1995. Descriptive Translation Studies and Beyond, Amsterdam: John Benjamins. Zabalbeascoa, P. 1994. “Factors In Dubbing Television Comedy”. In Perspectives. Studies in Translatology, 1994: 1, Copenhagen: Museum Tusculanum Press, 89-100. ----------. 1994. “Awareness in Translation”, Language Awareness Vol. 3: 3/4, 1994, Clevedon: Multilingual Matters.