On Semiotic Aspect of Translations

Boguslaw P. Lawendowski

A bundle of processes subsumed under the heading Translation is inherent to human and animal existence. Such an assumption can be substantiated only if the traditional concepts and units of translation are temporarily set aside in order to give room to a semiotic framework for the processes. To make the following exposition clear it seems necessary, however, to recapitulate as concisely as possible the past and the present developments in the theory of translation, and to indicate unambiguously what notions are to be set aside.

One of the signals of the growing immensity of the body of writings on translation is the fact that authors of major works in the field decide to list only a few scores of bibliographical entries they deem essential to the study of the field, leaving it to the student to discover other related materials for himself (cf. Steiner, 1975). Only a little more than ten years ago one of the most significant publications in the field listed more than 2,000 entries ranging from one-page articles to vast treatises (Nida, 1964).

Let us imagine a determined and diligent analyst ready to read his way through the enormous pile of texts in search of some unequivocal account of the translation processes. The task upon its completion will not leave him empty-handed, to be sure, yet he may be confused by the cornucopia of attitudes, definitions, beliefs, theories, opinions, and more or less relevant comments. It will transpire from the readings that the multitude of authors agree, as it were, to disagree.

This chapter was written while the author was an American Council of Learned Societies Fellow for 1975-76 affiliated with the Research Center for Language and Semiotic Studies at Indiana University, Bloomington, Indiana.

What are the reasons for this perhaps unhappy state of affairs? An answer to the question lies within the area of interaction between different anthroposemiotic systems as well as the interface between the anthropoand zoosemiotic systems. The latter ones go beyond the scope of the present paper but will eventually be included in other writings on the semiotic framework of translation. The reader is kindly referred to the extensive treatment of zoosemiotics in the works of Thomas A. Sebeok, and also to Sebeok (1975), in particular, since it presents a very lucid taxonomy of the six species of signs.

The whole field of translation viewed in retrospect presents itself first of all as a number of activities based on empeiria. In the chronologically initial period theõria of translation did not exist. The impact of this historical development cannot be overestimated, regardless of the fact that it was the way it was with the origins of all the essential building blocks of the human world.

What differentiates human communication from other human endeavors is its enormously dynamic and heuristic nature. This, in turn, creates a formidable obstacle on the way to an ultimate end: an adequate and universal register of units and their interaction within the communication processes used by man. The rift seen today between the empirists and the theorists of translation is due to ages of an absence of a unifying theory of communication and to the lack of tangible effects of linguistic theories of the past (as regards translation, of course). As a matter of fact, translations continue to be produced by experienced practitioners, or it would be better to say, are assigned to persons with generally greater empirical than theoretical backgrounds. Numerous instances of the real-life situations confirm the fact that even the most impressive theoretical apparatus at hand is a poor substitute for several years of intensive practice in the field. The situation is likely to prevail in years to come, and the former flood tide of interest in machine translation has ebbed considerably, indicating that in this particular contest between man and computer it is man who comes out a winner, at least in certain types of translation. Now, obviously the last comment could easily take us into another, long discussion of what computers are and are not able to perform. Since it is not the basic subject here let us only repeat that the problem in machine translation lies with software rather than with hardware, which is to say that we need not worry what the machine can do for us but what we can do for the machine.

It is not our intention to call for a fast change in the field in favor of practice more solidly based on theory. The purpose of the paper is to draw attention to the underlying aspect of all translation processes— semiotics: "the exchange of any messages whatever and of the systems of signs which underlie them" (Sebeok, 1975). The establishing of a semiotic framework for subsequent discussions of the translation processes may help to show them as operations infinitely more complex than the traditional linguistic, literary, philosophical, logical, or mathematical propositions are ready to admit. The semiotic framework may help to bring together the efforts of those analyzing and describing translation only within the confines of their own disciplines.

Unrewarding as they are, speculations and statements concerning the origins of language are continuously elaborated. Lieberman (1975) puts emphasis on the evolution of the human supralaryngeal vocal apparatus, unique among primates, which produced the particular form of human language. Hence, vocal signs seem to be of primary importance in the development of human culture. The same author points out, however, that "one of the great mysteries' of human evolution is the sudden acceleration in the rate of change of human culture that occurred between 40,000 and 30,000 years ago" (1975:180), evidenced in diversification of tools, appearance of new technological materials and techniques, as well as the coming of art. Accompanying the art is the first body of engravings, which indicates "the use of technology to expand human cognitive abilities. Orthographic systems can be viewed as devices that overcome the limits of human memory, whether the orthography is numerical, alphabetic, or syllabic." Lieberman refers to a recent work (Marshack, 1972) that reveals the existence of intricate notational systems well into the pre-history of man. He admits that "the exact nature of what was being recorded is not always clear. However, the startling thing is that notational systems were in use 30,000 years ago. Human morphology, cognitive ability, and language probably were fully evolved [italics mine]" (Lieberman, 1975:181).

This and similar statements point to the existence of diverse sign systems in some of the earliest stages of human culture. Each of the sign systems must have been born out of the need to communicate, thus the skill of either producing or interpreting signs is as old as the sign systems themselves. The production and the interpreting were, or, put better, have always been taking place at several levels: (1) Personal; (2) Interpersonal, which falls into (2a) Intra-tribal or Intra-societal and (2b) Inter-tribal or Inter-societal. A parallel set of terms could be (3a) Intra-cultural and (3b) Inter-cultural.

The evolution of the orthographic systems provided a dual system of language signs, which stratified the society into illiterates and literates, that is to say, a group of individuals able to exchange messages only by means of a single sign system in language—speech (plus some 'nonverbal codes'), and a group able to use freely the two linguistic sign systems. Social, cultural, political, and other implications of this stratification need not be elaborated here; suffice it to mention the role of the orthographic systems in religion. Versatility and economy of the writing systems (though, as we know, some are more economical than others) account for their dominance among other anthroposemiotic systems used in the signaling behavior of man. The position of the linguistic sign systems is unchallenged and will remain so, no doubt. But at the same time it is worth pointing out that the importance of language overwhelmed if not stifled attempts to consider communication as a result of messages being exchanged via several channels simultaneously, which eventually led to the establishing of the term translation as representing the transfer of meaning' from one set of language signs to another set of language signs at what we call here the 'inter-societal' (2b) level. In other words, translation could mean only one thing, as Catford (1965) put it: "an operation performed on languages: a process of substituting a text in one language for a text in another." (It is interesting to note that Catford refers to Roman Jakobson in a single footnote on p.74 of his work but passes in silence over Jakobson's famous article on linguistic aspects of translation [Jakobson, 1959], which should be regarded as the first step toward a semiotic concept of translation.)

Jakobsonian 'intralingual', 'interlingual', and 'intersemiotic' transiation processes bring us back again to the developments in the past history of translation. But first it seems in order to discuss briefly the two terms interpreter and translator. Both persons are engaged in the exchange of messages by means of transferring 'meaning' from one set of language signs (Source Language: SL) to another set of language signs (Target Language: TL). The interpreter, however, works with speech units while the translator does it with units of the writing systems. The enormous body of writings on translation is primarily concerned with the translator and the written texts. To be sure, all available translation theories deal exclusively with written languages, while interpreting is usually mentioned as a Very difficult type of oral translation, better left to practitioners only. On the other hand, to do justice to the handful of authors solely concerned with interpreting, we must say that since World War II a few books as well as a number of technical articles have been published, as, for example, Cary (1956), Min'iar-Beloruchev (1969), Herbert (1965), Seleskovitch (1968), and others. The growing interest in interpreting, techniques of which are discussed later in the paper, stems from the development of highly advanced electronic devices; for example, one enables a group of 1,200 people to communicate simultaneously in six, seven, or more languages.

Chronologically, the interlingual interpreter came first, and in all likelihood, when acting at some inter-tribal contacts he had to be familiar with or take quick guesses at signals from the opposite party, which belonged to what semiotic literature calls 'non-verbal codes': gestures, clothes, symbolic gifts, and others. He had to have sufficient knowledge to distinguish between signals, symptoms, icons, indexes, symbols, and names (cf. Sebeok, 1975), as different signs, although the knowledge must have been intuitive and confirmed only empirically. Consequently, the interpreter combined in his act information conveyed to him through a whole array of channels. This process continues today in almost the same way on such occasions as diplomatic visits. It is of considerable assistance to the interpreter that he is receiving information in forms other than a string of spoken words. That helps him do a better job of exchanging the messages, but his work is affected by the need to act fast.

The translator, on the other hand, has a lot of time to think over the text in SL, consult reference books, specialists, etc., but no other channel of information other than the written or printed text is available to him (unless there is a direct contact between the translator and the author of the text, in which case the translation may be done more accurately). The total effect of his work depends basically on his command of the SL and the TL, which implies mastery of both cultures.

The entire body of texts on translation, enormous as it is and largely concerned with the work of the interlingual translator, can be roughly divided into three categories: rhapsodic, technical, and theoretical. The rhapsodic writings embrace a considerable part of texts concerned with translation of the literary materials: prose, poetry, drama. A typical piece from this category lays stress on the difficulty of performing an act of literary translation by reiterating statements about the elusiveness of the language of belles lettres versus the simplicity and unambiguity of the language of science and engineering. Either bluntly or in a disguise of a more or less effective figure of speech the message that comes across from the author of such a piece is that the language of literary arts does not lend itself to systematic examinations, consequently only the gifted have the right to talk about it; implicite, he is one of the chosen few. Often texts of this kind are sprinkled with verbal flowers of praise for colleagues working in the same field. A common feature is a series of quotations from the translations, yet only some authors carry out a step־by־step analysis of the features that make a given translation good or bad. Most often the quotations are there for the friendly reader to admire, contemplate, and draw conclusions for himself. This category includes critical reviews of literary translations. More often than not the reviewer will gloss over the fundamental issue of whether a replacement of the SL material by equivalent textual material in TL took place. Lexical analysis may appear in some limited form if the translator happened to make a few outrageous errors. This type of reviewing may be due to the limited space in journals, but also, which is not an uncommon event, unfamiliarity of the reviewer with the language of the original. Generally, texts in this category take their origins in emotional rather than rational attitudes of their authors.

The technical texts often present themselves as more or less detailed taxonomies of the procedures the translator must follow to arrive at a successful end. Titles such as "Twenty-Three Restricted Rules of Translation" (Newmark, 1973), "The Principles of Translation" (Savory, 1968), or "The Lively Conventions of Translation" (Arrowsmith, 1961) abound here. Chapters on translation through the ages are usually ineluded, together with some paradoxes that result from juxtaposition of the views on translation expressed by different authors at different times. The best sample of it the reader can find in Savory (1968:50). The authors often provide ample bibliographies and discuss views of other authors writing in the same way. They do not try to build or fortify the existent mystique of the translation of poetry, for example. On the contrary, they tend to admit that an act of translation is only secondary to a genuine creative effort of a writers mind. Nevertheless, they can hardly put up with an idea that translating may be regarded as a field of systematic scholarly analysis: Cary (1965): "C'est qu'il n'existe pas de traduction dans l'abstrait"; or Newmark (1973), shedding the traditional moderation of a British subject: "purely theoretical treatises on translation are even less profitable than most purely theoretical treatises," and later on: "in spite of the claims of Nida and the Leipzig translation school . . . there is no such thing as a science of translation, and never will he [italics mine]." A number of works in this category raise the issue of untranslatability, although this problem is actually dealt with more extensively in the theoretical category. Jakobson's contention that whatever is expressible in one language can be translated into another, or in other words, the entire body of human knowledge may be expressed in any language, is not shared by many authors (cf. Werner Winter's "Impossibilities of Translation" in Arrowsmith, 1961). W. V. O. Quine was skeptical, only a year after Jakobson's article, maintaining that if anyone tried to translate "Neutrinos lack mass" into jungle language, he would have to coin words or distort the usage of old ones: "We may expect him to plead in extenuation that the natives lack the requisite concepts; also that they know too little physics. And he is right, except for the hint of there being some free-floating, linguistically neutral meaning which we capture, in 'Neutrinos lack mass', and the native cannot" Quine (1960:76).

It seems that Jakobson's assumption is valid if we recognize that interlingual translations are divided into different groups each of which has its own criteria of accuracy, quality, and style. If the purpose of a translation is, for example, to give the reader a good idea about some referents that are nonexistent in the TL, without the need to care about stylistic effects of the translator's work, then certainly the coining and distorting of terms will take place for the purpose of approximating the concepts expressed in the SL. The category of theoretical studies on translation is as diversified as its composition. The number of works entirely devoted to theoretical aspects of translation is small. Nida (1964) and Nida and Taber (1969) are the first and foremost publications arguing strongly in favor of systematic studies on translation processes. Catford (1965) applies the Firth-Halliday linguistics to translation, yet on a closer look it appears to be divorced from reality, at least some categories and taxonomies proposed by the author are bound to remain as Catford's idiosyncratic inventions. Of the most recent major works, we should again mention Steiner (1975), which, long as it is and over burdened with verbiage to boot, brings hardly anything new into the theoretical area of translation. If anything, Steiner seems to have a rather gloomy vision of the possibilities of translation.

The problem of untranslatability accounts for the diversity of the category. The quantity of articles, chapters, sub-chapters, and the like concerned with the impossibilities in the field is enormous, much too big even to begin listing them in the present paper. Writers from a very broad spectrum of disciplines continue to discourage any hopes for translatability. For example, Quine (1960) brings in his thesis of the indeterminacy of translation, or Wierzbicka (1975) informs the reader that there is more to the understanding of the process of translation than many a simple mind can imagine, maintaining that a sentence such as "She has only one parent" does not necessarily correspond to "She has only one mother or father" but rather to something worded in the following way: ("She has only one) parent" equals ("There is only one person who can be thought of as a grown-up member of a group one member of which is someone who is not grown-up, another member of which is this someone's father or mother"). It is a comforting observation, however, to see publications that argue with the 'untranslatability' attitude, like, for instance, Darmstadter (1967).

Generally speaking, texts falling into this category either evade or treat very superficially the practical matters (not necessarily because by definition they are not concerned with them but because many of the writers are not familiar enough with the kind of work the translator or interpreter is doing, and, applying the principle pars pro toto they question the validity of the entire endeavor). In the view of this author, the statement made by Searle (1969:11) is the one to be adopted by anyone elaborating on translation processes :

Searle speaks only of language. Yet, it may be worth examining how much of what he says pertains to other sign systems that we know as native 'users' and makers' of a given culture. Some are more, others are less, familiar, so classificatory work will be inescapable (like, for example, the application of the six species of signs, as described by Sebeok, 1975). Perhaps the six species will appear insufficient for our purposes. A future theory of translation will have a semiotic framework embracing not only language but other anthropo- and zoosemiotic systems, although the priority will no doubt rest with language. In order to work out such a framework, which would unify the presently dispersed and individualistic approaches to translation, interdisciplinary cooperation is necessary.

The 'semiotic' category of writings on translation has been limited so far to only two works: Jakobson (1959) and Ludskanov (1975). Jakobson's remarks concerning intersemiotic processes were not treated extensively in his article because its main thrust was of linguistic nature, yet it makes him the originator of the semiotic approach to translation. Ludskanov's was the first explicit statement in favor of regarding the field of translation as an integral part of semiotic studies, although he points out that the theory of semiotics does not account for 'semiotic transformations' though "such transformations certainly exist." To remedy the situation a definition of semiotic transformations is offered: "Semiotic transformations (T^s) are replacements of the signs encoding a message by signs of another code, preserving (so far as possible in the face of entropy) invariant information with respect to a given system of reference." It clearly indicates Ludskanov's greater preoccupation with the theory of information than with that of semiotics, in fact, but in the course of time narrower and more adequate definitions will be worked out. A great merit of Ludskanov's paper is the stress put on the existence of a number of relatively new types of translation work that cannot rely on the traditional approaches : film and television dubbing or subtitles, interpreting conferences, artificial computer languages (why he should consider translation of children's books as a new field for translators is not explained!).

Fig. 1 is the matrix covering practically all anthroposemiotic systems as used in the interpersonal or intersocietal exchange of messages. It certainly offers a good start toward the ultimate: a semiotic theory of translation, even though it is clear that Ludskanov's starting point for his semiotic approach is information theory. That it may not necessarily be the best direction for us to follow (information theory, that is) is evidenced by failure of computers to prove themselves to be of greater value to a human translator than just a group of sophisticated and fastworking electronic devices usable only as aids (cf. Rubenstein, 1974: 2712). As we have mentioned earlier, the problem may actually be due more to man's failure to provide appropriate software, so it may develop in time that a semiotic framework for translation could bring the computer back into the game.

One aspect of Ludskanov's matrix, however, seems to qualify it as a 'semiolinguistic' rather than 'semiotic' (the notion of 'semiolinguistic' exchange of messages is discussed further in this paper).

The interpreter works in ways distinctly different from those of the translator. The latter, as was mentioned earlier, usually works without haste, and he is not pressed by the situation to complete his task in one fell swoop. He may do it in a discontinuous way, shelving the text for a while and returning to it at will. Works of literary art sometimes call for years of effort (the Polish version of Joyce's Ulysses took the translator more than ten years to complete).

The interpreter, on the other hand, has "less time than no time at all," as a Thurberian character would say. There are two basic techniques of interpreting: consecutive and simultaneous. Mixed techniques are used also in ad hoc situations. Detailed descriptions of each technique can be found in Seleskovitch (1968), but for our present purposes let us outline them in very broad terms.

In the course of the consecutive process the interpreter becomes a link between the parties involved, that is, speakers of at least two mutually unintelligible languages who intend to exchange oral messages. The interpreter listens to a message, or a fragment of it, in the SL, and when the speaker stops his delivery the interpreter renders the message in the TL. The length of text that the interpreter takes in at a time depends on his skill, primarily on the efficiency of his memory. The longer his memory span the fewer the intervals in the exchange of messages. The famed interpreters of the former League of Nations are described as being able to take in up to thirty minutes or more of text at one time. Since individuals with such unique memories are not numerous, a special notation for consecutive interpreters came into use.

The interpreters' notation is designed to register the order of ideas in a given passage of the message. The actual wording is much less important; therefore shorthand systems are not used extensively although some elements of shorthand constitute a part of the interpreter's notation. Abbreviation, acronyms, and ideographic signs are combined into one. Examples of such notations are given by Rozan (1965), Hoof (1963), Min'iar-Beloruchev (1969). The Russian system, as given by Beloruchev, relies to a large extent on signs or modification of signs used in topographical cartography: battle, fighting, war; factory, industry, etc.; bilateral conference; riots, revolution, etc., to give only a small sample; marks the past; the future, and stands for now, present time. By recording a string of abbreviations, acronyms, and ideographic signs the interpreter helps his memory to recreate the message in the TL keeping the same "train of thought," so to speak. His command of the two languages involved should provide him the comfort of doing without the recording of the actual words used by the speaker. It must be stressed that since spoken messages are wrapped in a bundle of accompanying non-verbal messages, the interpreter has a difficult decision to make each time he does his job: how much of the non-verbal message should he convey? Ideally, of course, he should be as neutral or nonpartisan as possible since his sole preoccupation should be to provide the closest possible equivalent of the SL material without unnecessary embellishments or reductions. This is simply not feasible because as a human being he utters his text enveloped in what has come to be known as paralinguistic' or 'rosodie' features, and his face, for example, may convey certain quite unintended signals. Should he identify himself with the speaker and his style of delivery? Should he perform with as much detachment as possible? Such problems pertain to non-verbal codes more than to anything else. Admittedly, interpreters working even at the highest political levels are no more than human devices for accomplishing the necessary exchange of messages. Quite often, if not regularly, they hide behind the personalities they are working for at the moment, so that a picture of two heads of state, for instance, may show the two politicians smiling broadly at each other and giving the impression that their contact is as personal as can be. Aspects of such situations have received a very interesting treatment by Goffman (1966).

It is the work of the consecutive interpreter and the notation serving him as a mnemonic device that deserve an extensive semiotic analysis. A study of the existent notations and the working out of a single international system of ideographic notation offer an excellent opportunity for a monographic paper at least.

The work of a simultaneous interpreter is considerably different. He has no contact with the orator other than through a set of earphones. He has a very limited possibility of influencing the delivery of a given speech. More often than not, if the conference hall is equipped properly, the speakers do not even know and cannot see the interpreters in their booths. In some cases the interpreter can see the speaker only via closedcircuit television, and the speaker does not monitor the translation, so that while addressing his audience he assumes that his text is being interpreted. The simultaneous technique is relatively new in the field; it is made possible largely through the latest developments in electronics, such as miniaturization by the introduction of integrated circuits. Not only is the number of interpreters capable of performing this way limited but also many speakers at conferences provided with simultaneous interpreting do not quite understand the specificity of the technique. If an American whose training includes courses in public speaking, which invariably tell one to begin an address with a joke to set the mood on a positive if not friendly note, happens to be the first speaker of the day, when neither the audience nor the interpreters have yet "tuned in" properly, the poor orator is more often than not astonished at the listeners' different sense of humor, which forbids them to appreciate his wellchosen specimen.

The name of the technique is, of course, a nice simplification. It is hard to imagine two individuals generating simultaneously two sets of signs that would bear very similar messages. The interpreter implicite follows the delivery in the SL. Now it is true that many public statements are filled with clichés or have stereotyped constructions that may help the interpreter guess what forthcoming portions of the message will contain, but he cannot always be totally sure of that. Even if typed or written texts are provided by the speaker as the graphic versions of his address the interpreter may not rely entirely on the speaker's word; during some extended public speeches the speaker may mistakenly miss a page or two of the pre-edited text, and, consequently, the interpreter who does not notice the omission in time will depart from the SL version considerably. An experienced interpreter can mend the situation, but this is not of immediate significance to the present topic.

Some features of the simultaneous technique deserve a closer semiotic look. The speaker and the listener are totally aware at the outset that their communication is accomplished through a third, invisible and often multipersonal, party. With the passing of time during a given speech, however, it is up to the interpreter to develop in the mind of the listener a notion of directness—that he is listening to the speaker himself. Though it is difficult to imagine it is nonetheless performed by some of the more gifted interpreters, and then, of course, a situation appears that is essential to all translation: If a written code, a TL code, that is, may impress itself on the reader's mind to the point that he stops thinking in terms of the SL and TL codes and only considers the TL while reading a given message, it is actually much easier for the interpreter to impress the listener in the same way since he has at his disposal the entire apparatus of vocal devices and features, which carry perhaps more emotional than cognitive values but which, nevertheless, are regarded highly as identifiers of belonging to the same culture, subculture, social subgroup, family, etc. An act of communication between the speaker and the listener(s) is accomplished by a human being (or several of them acting in a sequence) who uses electronic gadgetry, such as microphones, earphones, amplifiers, loudspeakers, batteries of push buttons, etc., to perform his function as a go-between for two parties using different languages. On account of his reliance on the electronic devices we may say that the ratio of masking noise, which impedes straightforward comprehension of both the SL and the TL messages, may be regarded as a more formidable obstacle than, for example, garbled handwriting or typescript.

As we have already pointed out, the time factor in this type of translation is of the utmost importance. In fact, potential interpreters unable to operate under such difficult conditions do not qualify for the job, as Johansson (1975) indirectly suggests in his concise paper: "a measure of a subject's ability to receive messages under masking noise is a good indication of overall proficiency, and . . . it might be possible to define a subject's proficiency by reference to a signal-to-noise ratio, i.e. the maximum noise level at which a subject can function efficiently." We mentioned above that several interpreters can act sequentially, translating different sections of the SL material. Instances of this type are most easily encountered at large international conferences where a single speaker, who happens to be a prominent politician, may continue single-handedly for several hours. This raises a minor but valid question: how does the switching of interpreters affect the communication? In addition, a word must be added about the practice, which is not openly discussed in the interpreters' milieu as it is regarded as a sort of disgrace, of interpreting via a "relay" system: an exotic, so to speak, language used at a conference in Europe may first be interpreted into one that is a standard conference language and then retranslated into other Ianguages used at the given event. Such practices are certainly not supported by the statutes and policies of the International Association of Conference Interpreters (A.I.I.C., Paris) but everyday life situations often invalidate statutes. The loss of information in the course of such a "relay" process is also an interesting area for investigation. The fact that interpreters working at international conferences routinely switch every 30 to 45 minutes and that the listener is simply forced to combine his visual observations of the speaker with the vocal impressions of one or more hidden interpreters is another field for psycholinguistic and semiotic analyses. The disconnection between the originator of a sound message and his (or her) image while speaking is a standard device in literary, folk, and religious tales. The absence of the visual image combined with usually pathetic tone of the pronouncement builds a special communicational effect. Is it present in interpreting at conferences? The speakers of a given language have certain stereotype linkages built into their communicational circuitry. How does this factor affect the exchange of messages in simultaneous interpreting? This and related problems deserve systematic investigations if we want to account for the multiple aspects of translation as practiced today.

The matter of the proper combination with visual images of voice set, speech rate, and other paralinguistic or prosodie features is fundamentally important in the area of translation for the film industry. Two techniques are used with different effectiveness. The first one is that of providing frames with subtitles in the TL, leaving the entire SL phonic substance as is. This particular technique has only one interesting, from our point of view, communicational aspect. The frame makes it imperative for the translator to limit his text to a prescribed length, for where there are long passages of spoken text in the original dialogue a rendition of the complete text might cover the entire frame. The translator must decide what parts of such long utterances can be left out without hurting the verbal substance of the film.

A more complicated and hybridic operation occurs during dubbing, another technique of translating the original dialogue of the film into a TL. The peculiarity of the process lies in the need to synchronize with the lip movement of the film heroes, who, in most cases, speak their native tongues, although the film-making of the ^,60s and ^,70s introduced additional authenticity to certain sequences involving speakers of other languages (cf. scenes with German, Japanese, or Russian soldiers and officers in movies about the Second World War, to give only one example). In such instances the English version of the film is provided with appropriate subtitles. The authors of the dubbed text take into account a peculiar feature of communication: the viewer wants to see compatibility between sounds and lip movement, otherwise he immediately detects "foreignness" in the dialogue, which in turn leads to his distraction, discomfiture, and what have you, consequently ruining the job of the director. It is worth noting at this point, however, that certain categories of films tend to have a very fixed vocal-visual image, often because of the profusion of the kind: the Western, with its cowboys, mountains, colts, etc., is so strongly linked with American English that dubbing in the German "Fräulein Brown" for "Miss Brown," for example, would make the viewer feel very much ill-at-ease. This author circulated a questionnaire among a number of Polish students at Warsaw University in an attempt to determine whether dubbing was a favored technique.

Most of the answers were to the contrary: most of them stressed the "incompatibility," as the respondents called it, of scenes filmed in the American Southwest, where cowboys, Indians, and lone hunters played major roles, with dialogues in Polish; or of takes of the streets of Paris with Jean Gabin swearing in Warsaw Cockney! The students agreed, on the other hand, that films composed largely of close-ups of the actors or that rely heavily on facial gestures (e.g., Twelve Angry Men with Henry Fonda) do call for dubbing. In most cases they recognized the detrimental effect subtitles have on the reception of the visual signs of these movies, but they still preferred that hard way, seeing in it more authenticity and more directness. It is true also that some countries dub every film they import, but their reasons may be different, illiteracy of their audiences being only one of them, perhaps. A British translator writes about the dubbing of films in a way that certainly deserves a quotation:

Another type of interpreting that presents specific difficulties is the simultaneous oral delivery of the already translated text. Jean Ure (1964) says that in such cases the interpreter must be a particularly alert person since the laughter or the applause must come to the same signal, and he quotes translations of plays read aloud by interpreters and relayed to the audience through earphones, as, for example, in the Jewish theater in Bucharest (using Yiddish). To the best knowledge of the present author, similar techniques are used in the Jewish theater in Warsaw. Ure mentions also that the Chinese theaters and opera houses project a written translation by means of a magic lantern onto a screen beside the stage, which makes it a part of the subtitling category, but technically different from that of the film industry.

There are also such rarely practiced forms of interpreting as the whispered simultaneous or consecutive, or the press agency teletyperoom translation/interpreting (a fast teletyped text in the SL is immediately translated orally to a typist, who produces the typewritten TL version), and others. It was not my basic objective to discuss in detail all practiced forms of translation and interpreting, consequently, the reader interested in technicalities is referred to a diversity of papers scattered throughout the volumes of Babel.

An anecdotal interlude is perhaps in order here. A few years ago I was asked during a class if I saw any resemblance between Sir Alec Guinness, the actor, and a horse. Since no teacher likes to be made a fool of in front of his students I dragged on with a direct answer while forcing my mind to work furiously in search of all possible associations between a man and a horse. As well-worn phrases indicate, man likes to involve animals or their features when talking about other men. Entire populations will ascribe animalistic tendencies or characteristics to their neighbors or even to peoples from far away. With all respect to our British friends I must say that the Poles tend to see the lower jawbones of some Englishmen as much more pronounced than their own and, therefore, may refer to the innocent British subjects as people with 'a horse's jaw' (końska szczęka). I thought that the odd question stemmed from such an attitude but I found out later that it was a case of a poor translation of a film title. A Joyce Cary novel (The Horse's Mouth) and a film based on it had identical titles. The Polish version of the novel had a proper idiomatic translation of the title (Z pierwszej ręki) but the film appeared as Koński Pysk (A Horse's Mouth), which made many a movie-goer in Poland wonder how much equine beauty the leading man had.

Can we translate messages in Van Gogh's pictures, violin concertos, or Navajo rug designs? First, let us not confuse the relatively narrow concept of 'intersemiotic' processes introduced by Jakobson (1959) with the much broader framework of semiotics known to us today. He indicated the interaction between non-verbal signs and language in a particular way. Things such as traffic signs, for example, carry conventional messages adopted by members of a given community but if a sign turns out to be easily interpretable in more than one way attempts are made to replace it with a univocal, as it were, message. Therefore even though formally such signs are not verbal they are strictly an offshoot of language. In our understanding, the Jakobsonian 'intersemiotic' processes are not really intersemiotic since they automatically involve language. They are, for want of a better term, 'semio-linguistic' processes, as is the case with all instances when non-verbal messages are processed into verbal, regardless of whether it means vocal or graphic externalization or what we know as 'inner speech'.

A process closer to the intersemiotic exchange should embrace direct interaction of non-verbal elements, without the go-between of language. Music is perhaps the best example here. The highly subjective personal reception and internal interpretation of music does not require verbalization in any form (although one may indulge in such an activity trying to prove, for instance, that indeed there are dozens of shepherds in The Pastoral Symphony). The fact that music signals may by-pass the language channel and go through directly to our inner emotive and cognitive circuits makes them possess the fundamental feature of what we propose to call 'the semiotic processing', rather than 'intersemiotic' translation, transmutation, etc. Certainly we cannot 'translate' Van Gogh's powerful messages, but at the same time, while receiving them visually, we 'process' them in a number of ways, one of which can, of course, be a verbal externalization of our personal impressions and emotional reactions or an extension of our cognitive capabilities.

The semiotic framework for translation theory proposed in this paper assumed that man and animals are immersed in a sea of signals. The processing of signals starts at birth and ends with death while going on simultaneously at many levels, some of which we are familiar with and others which still escape our grasp. Semiotics offers an opportunity to work out tenets of an approach that could be accommodated interdisciplinarily. This is particularly clear in the field of translation, where so far randomness, subjectivity, and parochialism dominate.

The inevitable shortcomings of the presentation derive primarily from the absence of opinion of others working in the field. The views expressed here are no more than a brief attempt at stimulating interest in one of the fundamental activities of man and animals: the exchange of messages.