THE THEORY OF TRANSLATION
4. Translation theory and text linguistics
Back to the text! This is how one could characterise the translation theory of the 1980s. Thinking about the text has been a part of thinking about language for two thousand years, but that thinking did not give rise to text linguistics as an inde
pendent field of study, in the same way as thinking about translation did not add up to and was not called translation theory. It is worth remembering, however, that the idea that the text plays a central role in translation, however novel the topic may sound, has been present for many centuries.
4.1. Text-centredness in translation
Translators have never translated anything else but texts, and the outcome of translating has never been anything else but text. The process of translation has never consisted of anything else but the comprehension of the original and the creation of the translated text.
If, however, text-centredness has always been part of the thinking about trans
lation, then why do we say that translation theory has returned to, or has found its way back to the text? When did they separate? Paradoxically, it was the linguistic theory of translation that was responsible for this separation.
4.2. Disregarding the text-level
The linguistic theory of translation achieved the status of an independent disci
pline in the 1950s and 1960s owing to the fact that, in addition to studying liter
ary texts, it extended the scope of its investigations to scientific, technical and oth
er non-literary texts, using the tools of linguistics. Since linguistics at that time was focused on the sentence, the use of linguistics in translation theory shifted the focus of attention from text to sentence. Equivalences between source language and target language were studied on the sentence level and below, on the level of words, phrases, and grammatical structures.
The first sign of this approach is seen in Retsker’s famous article O zakonomernih sootvetstviyah pri perevode na rodnoy yazik (On regular correspondences in transla
tion into the mother tongue), published in 1950, three years before the appear
ance Fedorov’s fundamental work on translation theory. Retsker’s study argues for the existence of “regular correspondences” (zakonomernie sootvetstviya), which apply irrespective of the situation, the context, or the text. This accounts for the fact that equivalences between the words and phrases of the source language and the target language can be described, systematised, and classified. Retsker distin
guishes three kinds of correspondences: “constant correspondences” (postoyanniye
4. Translation theory and text linguistics
sootvetstviya), such as technical terms, geographical names, etc., “variant corre
spondences” (variantnie sootvetstviya), where the target language offers several possibilities and translators have to choose among them, and “occasional corre
spondences” (okkazionaVnie sootvetstviya), where the translator has to create cor
respondence on the basis of the context (Retsker 1950, 1974, Retsker in Zlateva 1993). At that time this approach was considered pioneering, since it drew atten
tion to the fact that translation, a process seemingly consisting of a series of sub
jective decisions, was in fact guided by objective rules.
Following the publication of Retsker’s and Fedorov’s research, a number of manuals appeared in the former Soviet Union, describing the lexical and gram
matical equivalences characterising specific language pairs: e.g., Gak and L’vin 1970 (French-Russian), Roganova 1971 (German-Russian), Levitskaya and Fiterman 1973 (English-Russian). In retrospect, the limitations of such studies seem obvious. They considered solely the two linguistic systems involved in translation, and saw a translating problem in every single difference between the systems. Based on linguistic differences, they tried to discover the “rules” of trans
lation. The literature of this early period of translation theory predicts translation problems, e.g., in cases where a particular grammatical category is absent in one language and present in another (e.g., articles in Russian and English), or where the structure of semantic fields show differences.
For this reason many scholars argue that research on translation theory in the 1960s and 1970s has more to do with contrastive linguistics than with translation theory as such. Even so, however, this was an extremely productive period, with researchers accumulating large amounts of raw material that could later be used for text-centred investigations. We must add in fairness that many of these early works on translation theory contained, usually somewhere at the end of the book, when all the description and classification of lexical and grammatical transforma
tions had been presented, a chapter in which the author discussed what they called
“stylistic transformational strategies”. Here the authors cited examples demon
strating how translators’ decisions are influenced by the genre of the source lan
guage text or the particular functional style to which it belongs (Shveitser 1973, Retsker 1974).
4.3. Returning to the text
In the early 1970s, with the emergence of text linguistics, the need for a text lin
guistic approach was perceived in translation theory (although the declarations made to this effect did not lead to practical consequences at the time). Text lin
guists, in their turn, also made frequent references to the possibility of applying the results of text linguistics research in translation theory (cf. Dressier 1973: 113).
A good example of the fact that the mutual recognition of the relevance of text linguistics for translation theory had no direct consequences in translation research is provided by Nida and Taber’s book, The Theory and Practice of Translation (1969).
The authors compile a list of 8 universal features of text, and suggest that they should be taken into account in translation. It is well worth presenting these eight features, since they represent a research agenda as yet unfulfilled. In Nida and Taber’s view, “all languages do have certain important features which can be used,
and which in ‘effective’ communications are used, to mark the units larger than sentences:
(1) the marking of the beginning and end of the discourse (2) the marking of major internal transitions
(3) the marking of temporal relations between events
(4) the marking of spatial relations between events and objects (5) the marking of logical relations between events
(6) the identification of participants (7) highlighting, focus, emphasis, etc.
(8) author involvement (Nida and Taber 1969: 152)
This research agenda has retained its relevance since 1969. Although, as we shall see later, there have been important advances in the analysis of translation from a textual point of view, no consistent discourse level analysis of these language-pair- specific characteristics has been carried out so far.
In the book Perevod i lingvistika (Translation and linguistics), published in 1973, Shveitser states that “for translation theory it is not only the comparison of systems that is important, but also the comparison of the textual realisations of the systems’ differences” (Shveitser 1973: 14).
In 1975, Barkhudarov made the following statement at a conference on trans
lation theory at Maurice Thorez Foreign Language College: “In translating, trans
lators do not deal with abstract language systems or linguistic units, but with spe
cific texts. The creation of a scientific translation theory is only possible with the comparative analysis of source language and target language texts. In the opposite case we can only speak about contrastive linguistics and not translation theory”
(Barkhudarov 1975b: 6).
The same ideas were raised by speakers at the conference Kontrastive Linguistik und Übersetzungswissenschaft in Saarbrücken in 1978. Reinhart Hartmann spoke about the interaction between translation theory and contrastive textology (Hart
mann 1981: 200), while Gideon Toury emphasised the viability of a tridimension
al comparison, which, in addition to the comparative analysis of the source lan
guage and target language texts, also attributes importance to the analysis of the translated text as a special form of the target language text (Toury 1981: 257).
However, all of these studies, Nida and Taber’s 1969 book The Theory and Practice of Translation, Shveitser’s 1973 Perevod i lingvistika, Barkhudarov’s 1975a Jazik i perevod, and even Reinhart Hartmann’s or Gideon Toury’s works failed to put these thoughts into practice. In other words, there are no large-scale studies comparing long stretches of source language and target language texts, drawing conclusions from such extensive analyses.
In the late 1970s and early 1980s increasing numbers of studies applied the research methods of text linguistics to the analysis of translations. In this period, two main trends can be distinguished: Chernyakhovskaya (1976) approaches the text from the inside and concentrates on the way the differences in the internal structure of the text affect the solutions of translators, while Reiss (1971) looks at the text from the outside and focuses on how the various text types influence the translators’ task. Let us examine both trends in more detail.
4. Translation theory and text linguistics
4.4. The internal text structure approach
Naturally, research on the relationship between text coherence and translation was preceded by the study of text coherence independently of translation in the text linguistics of the 1970s. The analysis of the internal structure of continuous texts, the most important field of study within text linguistics from the very beginning (Isachenko 1965, Harweg 1968, Weinrich 1971, van Dijk 1975, Halliday and Hasan 1976, etc.), became popular with translation scholars as well.
The idea that in emotionally neutral, descriptive texts, sentences start with either generally known information or with information known from the rest of the text (theme, thematic part) and the second part of the sentence contains new information (rheme, rhematic part) is not new in linguistics (it goes back to the first half of the 20th century; see the works of the Prague circle). This is called the theme-rheme, or topic-comment (conceptual, cognitive or logical) structure of sentences. However, analysis of the text-organising role of this conceptual struc
turing in continuous texts and its linguistic realisations was a new area of research. The application of the results of contrastive text linguistic studies to the analysis of translation was also a new departure.
Leonora Chernyakhovskaya ’s Perevod i smislovaya struktúra (Translation and the structure of sense) published in 1976 (in Russian) can be regarded as the first study not only to declare, but also to consistently apply the text linguistic approach to translation theory.
In Chernyakhovskaya’s view, transformational strategies used in the course of translation are not motivated by the different lexical and grammatical systems of the two languages, but by the need to preserve the structure of sense. The sense structure of the sentence (smislovaya struktúra), consisting of the part that is already known (theme/topic) and the part that is new (rheme/comment), and the emphatic part within the new information (focus) are expressed by different means in every language. It is exacdy this structure of sense that must remain invariable in translation. Accordingly, every single translational operation should serve the preservation of the sense structure characteristic of the source language text.
Chernyakhovskaya’s book does not stop at declarations: to support her claims, she provides analyses in which her principles are applied consistently. She exam
ines patterns of sense structure in English sentences on the basis of a large num
ber of newspaper articles. She explores the means English uses to signal the theme (topic) and the rheme (comment) and analyses the possibilities of marking empha
sis (focus) within the rhematic part. Then she goes on to compare the different types of sense structures found in English sentences with their Russian equiva
lents to establish the translational operations that the translator must perform in the case of different sentence types to preserve their sense structure.
Although Chernyakhovskaya uses sentences to illustrate her ideas and not texts, we can still regard her work as an example of discourse-level translation analysis, since all the sentences are taken from the same continuous text. Both given infor
mation, i.e. information that is known from previous sentences and new informa
tion were identified on the basis of the places they occupy within the English text.
The Russian translations used in the investigation were not translations prepared
by the author, but revised and edited translations made by professional transla
tors, in which the editors’ corrections were motivated by the desire to restore the sense structure where it was distorted by the translators.
4.5.The text type-typology approach
The other text linguistic trend approached translation from the direction of text types. The most outstanding representative of this trend, Katharina Reiss (1971), made an attempt to develop a translation-focused text typology (Übersetzungsrel
evante Texttypologie).
A genre-based classification of texts to be translated had, of course, been attempt
ed earlier as well. Fedorov, in the last chapter of his book (1953) discusses the genre-related translation problems of three groups of texts: (1) news and reviews, official and technical documents, and scientific texts, (2) political texts, newspaper editorials, and speeches, (3) literary texts. Mounin (1967) lists seven groups: reli
gious texts, literary texts, poetry, children’s literature, stage texts, movie texts, and technical texts. In earlier works, however, classification of text types was consid
ered a minor issue, and no-one intended to describe the characteristics of the process of translation entirely on the basis of the features of the text to be translat
ed. This latter task was undertaken by Katharina Reiss.
In her book Möglichkeiten und Grenzen der Übersetzungskritik she distinguishes four basic text types. She argues that the classification of texts is determined by the role language plays in the given text. Karl Bühler (1934, 1965) identifies three basic functions of language: Darstellung (‘description’), Ausdruck (‘expres
sion’) and Appel (‘appeal’). In most texts, all three functions of language are present, but one of them is usually predominant. This provides a basis for Katha
rina Reiss to make a distinction between content-focused texts (inhaltsbetonte Texte), in which the descriptive function of language dominates, form-focused texts (formbetonte Texte), in which the expressive function of language domi
nates, and appeal-focused texts (appellbetonte Texte), in which the appeal func
tion of language dominates. Besides the three text types determined on the basis of the function of language present in them Reiss identifies a fourth text type as well which reaches the receptor not via printed media. She refers to this text type as audio-medial (audio-medialeTexte).
4.5.1. The translation of content-focused texts
Content-focused texts involve a number of text types: press releases, commentaries, news reports, users’ manuals, patent specifications, official documents, non-fic
tion, specialised books, essays, reports, etc. Since these texts, despite the some
times rigid formal conventions they observe, are focused on conveying informa
tion, the translator’s task is to transmit the source language content in full, using the most appropriate devices of the target language to make sure that the reader’s attention is not distracted from the content. In other words, the fact that the trans
lator has to concentrate primarily on transmitting information accurately does not mean that he/she does not have to be careful about matters of form, i.e. the appro
priate use of target language forms characterizing the given text type.
4. Translation theory and text linguistics
4.5.2. The translation of form-focused texts
In the case of form-focused texts, it is not or not merely what the author says that really matters, but also how he/she says it. This text type also contains various genres: literary prose (essays, biographies, belles-lettres), imaginative prose (anec
dotes, short stories, romances), and poetry in all its forms. Here, the main task of the translator is to reflect primarily the form and not the contents. This is no easy task, however: the form is closely tied to the source language and thus cannot be automatically transferred into the target language.
According to Reiss, in such cases the translator will not adopt the source lan
guage form, but will try to get inspiration, and stimulated by this inspiration he/
she will choose the target language form that best approaches the effect in the tar
get language reader that the source language form produced in the source lan
guage reader. Therefore, in the case of form-focused texts the task of the transla
tor is not to produce identical content, but to create formal analogy.
4.5.3. The translation of appeal-focused texts
The third text type identified by Katharina Reiss is the group of appeal-focused texts. In such texts, both the content and the form are intended to provoke a par
ticular reaction in the listener or reader. This text type may involve appeals for likes and dislikes, or for specific actions (e.g., shopping) or the ceasing of specific actions (e.g., smoking).
The most typical text types falling into this category are commercials, ads, texts related to missionary work, propaganda materials, etc. Here the aim of the translator is not to reflect the content or the form of the source language text, but to render its function, i.e. to make sure that the target language text will provoke the same response as the source language text. To achieve this aim, the translator may deviate from both the content and the form of the text. In the case of adver
tisements, for instance, the same product must be promoted using different hints and allusions in different countries in order to preserve the appeal function of the text in the target language as well.
4.5.4. The translation of audio-medial texts
In the fourth text type the message reaches the receiver (audience, listener) via a channel the characteristic features of which need to be taken into account. Such texts are the radio and television genres and theatre plays, from operettas to operas and from comedies to tragedies.
In the case of translating a libretto the translator must not insist on a faithful rendering of either the content or the form of the original libretto if in the target language it does not match the music’s melody and rhythm, and cannot be sung with ease. Or let us just take the example of dubbing movies. Here the translator should watch the actor’s lip movements, and instead of preserving the content and form the translator should make sure that the dubbing actor’s sentence ends exactly where the original actor closes his/her mouth. Thus, in the case of audio-medial texts, translators have to take into consideration the conditions of the transmitting channel.
4.6. The typology of specialised texts
Vannikov’s 1987 study is an interesting attempt to develop a detailed and multidi
mensional typology of specialised texts. Vannikov lists twelve features on the basis of which scientific and technical texts have to be characterised to provide suffi
cient guidance for translators.
(1) On the basis of linguistic organisation
1.1 Texts with a rigorous structure and with strict linguistic formulation, 1.2 Texts with a soft structure, allowing the translator greater variety
regarding linguistic formulation;
(2) On the basis of the functional style 2.1 Scientific texts,
2.2 Technical texts, 2.3 Official texts, 2.4 Legal texts, 2.5 Journalistic texts;
(3) On the basis of functional register
3.1 Scientific texts (3.1.1 Academic texts, 3.1.2 Texts with an educational purpose, 3.1.3 Encyclopaedic texts);
3.2 Technical texts (3.2.1 Technical descriptions, 3.2.2 Instructions, 3.2.3 Technical information);
3.3 Official texts (3.3.1 Official directions, 3.3.2 Management texts, 3.3.3 Official correspondence);
3.4 Legal texts (3.4.1 Technical documentation, 3.4.2 Descriptions of inventions, 3.4.3 Patent management texts);
3.5 Journalistic texts (3.5.1 Scientific journalistic texts 3.5.2 Popular science texts);
(4) On the basis of manner of expression 4.1 Narrative texts,
4.2 Descriptive texts, 4.3 Explanatory texts, 4.4 Argumentative texts;
(5) On the basis of logical content 5.1 Exposition/Discussion, 5.2 Justification,
5.3 Conclusion, 5.4 Definition;
(6) On the basis of subject-related contents 6.1 Texts in exact sciences,
(6) On the basis of subject-related contents 6.1 Texts in exact sciences,