THE THEORY OF TRANSLATION
6. The concept of equivalence in the theory of translationof translation
Research on equivalence, that is, on the “equal value” of the source language and target language text, is an important field of linguistic translation theory. Equiva
lence is a central category not only within the theory of translation but also within linguistics in general. No language system can be described without an attempt to identify the sense- and content-related equivalences of its linguistic units. When comparing different languages, the relations between the different forms are also described on the basis of sense- and content-related identity or equivalence.
6.1. The equivalence perception of readers, translators, and researchers
The situation is much more complex in the theory of translation, where languages are compared from the point of view of translation, since in this case, determining the presence or absence of equivalence, further “participants” must be taken into consideration. Besides the researcher and the translator, the receiver (the reader or the listener of the translated text) also has own points of view. The concept of equivalence is viewed differently by the reader or listener, who assume equiva
lence, by the translator, who creates equivalence, and by the researcher, who investigates equivalence (Albert 1988).
The reader’s attitude is the simplest one. He/she places confidence in the translator. When he/she is given a translation, he/she will consider it equivalent.
Readers do not have a conscious view about equivalence; if they do not like a piece of writing, be it scientific or literary, they will blame it on the author. When readers blame the translator, they usually say that the translator mistranslated cer
tain words. Translators have a more or less conscious idea about equivalence.
Paradoxically, the quality of translation, the success or failure of the translator in creating equivalence is not closely related to the extent to which his/her ideas about equivalence are conscious or instinctive.
The problem of equivalence is treated as an issue of the greatest complexity in the work of researchers of translation. In contrast to the more or less instinctive, but still “evident” equivalence perception of readers and translators, researchers of translation have formed various less self-evident views about the essence of equiva
lence.
6.2. Approaches to equivalence
Some researchers consider equivalence as a precondition of translation, which distinguishes it from other forms of transformations into a foreign language, such as adaptation, abridgment, summary, etc. They include equivalence in the defini
6. The concept of equivalence in the theory of translation
tion of translation as a requirement or precondition (translation = replacement of the source language text by the target language equivalent). In this approach equivalence does not have subtypes or degrees in translation. If only those target language texts can be considered translations, which are equivalent with the source language text, then all translations are equivalent, and thus further research on equivalence is unnecessary.
The other approach starts out from the idea that equivalence is never complete.
The target language text intended as equivalent by the translator and accepted as equivalent by the reader is identical with the original text only from certain (for
mal, situational, contextual, communicative, etc.) aspects, and these are the aspects that have to be revealed and systematised. This approach is apparently more fruit
ful. Translating practice provides researchers with vast amount of data to explore the various types and degrees of equivalence.
Within this approach, two trends may be distinguished. One of these is the nor
mative view, which prescribes what the translator has to do to produce an equiva
lent translation; what it is that he/she has to definitely preserve from the original text and what it is that he/she can sacrifice (Fedorov 1953, Retsker 1974, Barkhu
darov 1975, Jäger 1975). The other trend is the descriptive view, which describes, on the basis of the analysis of numerous translating facts, how translators create equiva
lence; what it is that they have preserved from the original text and what it is that has been sacrificed (Revzin and Rozentsveig 1964, Catford 1965, Komissarov 1980).
A third approach to research on equivalence starts out from the idea that no identical equivalence requirements can be established for the translator of a users’
manual and that of a movie script or lyrical poem. In other words, the number of text types determines the number of equivalence types possible (Reiss 1971).
The present chapter excludes the investigation of the concept of equivalence from the point of view of the reader and translator (for the first see Popovich 1975, for the latter see Albert 1988). It will focus on the views of translation scholars who do not regard equivalence as a general precondition of translations, but attempt to describe various types of equivalences.
6.3. Catford’s view on equivalence
We find one of the first definitions of equivalence in Catford’s classic work A Lin
guistic Theory of Translation (1965). He distinguishes between "formal correspon
dence” and "textual equivalent”. Formal correspondent is any "TL category, which may be said to occupy, as nearly as possible, the ‘same’ place in the econo
my of theTL as the given SL category occupies in the SL” (1965: 32). His defini
tion of textual equivalence is rather vague: "any TL form (text or portion of text) which is observed to be equivalent of a given SL form (text or portion of text)”
(1965:27).
Catford continues by discussing the nature of formal correspondence, which can be only approximate. Because of the differences between the systems of lan
guages, lexical units and grammatical structures almost never occupy the same place within the hierarchy of the two languages. He cites English and French prepo
sitions as examples. In both languages they appear together with nominal groups in the structure of adverbial phrases, and function as either qualifiers in nominal
group structure (door of the house - la porte de la maison) or as adjuncts in clause structure. If, however, we want to move one step forward to show the equivalence of adverbial phrases containing prepositions, it can only be done on the basis of textual equivalence (1965: 33).
As Catford’s further examples refer to the establishment of formal correspon
dences between English and Kabardian, let us continue with examples of Hungarian and Indo-European languages. System identity would be the condition of formal correspondence in the case of lexical units as well. The English word brother can
not be a formal correspondent of the Hungarian word fivér, because Hungarian, in contrast to English makes a distinction between an elder brother (Hung: báty) and a younger brother (Hung: öcs). The Hungarian or English nagymama!grand
mother cannot be the formal correspondent of the Russian babuska ('grandmoth
er'), because in Hungarian and English beside nagymama!grandmother and nagy
papa! grandfather there is also the collective lexeme nagyszülők !grandparents, which does not exist in Russian. The German blau cannot be the formal correspondent of the Russian siniy, because besides siniy ('dark blue’) there is also the word gol- uboy (‘light blue’).
Catford’s example for the lack of formal equivalence is the English word yes, and the Japanese word hat, because Japanese hai, besides meaning jœs, also expresses a reinforcement to a negative question. The English yes fails to qualify as the formal correspondent of the French oui as well, because yes belongs to a binary system (yes-no) y while oui is part of a three-component system (oui-si-non).
Looking at these examples, the possibility of formal correspondence may be almost completely excluded from translation, because the identity of place in the system hierarchy only applies to international terms.
However, what Catford says about textual equivalence is, as mentioned before, rather vague: “any TL form (text or portion of text) which is observed to be equiva
lent of a given SL form (text or portion of text)”(1965: 27). He further specifies this as follows: “SL and TL texts or items are translational equivalents when they are interchangable in a given situation” (1965: 49) (author’s italics).
Catford cites the following example to illustrate this statement: In an imaginary situation a girl walks in and says: I have arrived. The situation has numerous ele
ments: the place, the time, the girl’s name, age, height, weight, colour of her eyes and hair, her clothes, profession, religion, relationships to other people; the num
ber and nature of her audience, and so on. Only very few of these are linguistically relevant, very few are built into the sentence I have arrived. These are the follow
ing: (1) the speaker (7 and not you or he) y (2) she has arrived and not left (<arrive and not leave) y (3) it is about an event that has already happened and not some
thing that will happen (have arrived and not will arrive)y (4) the prior event is linked to the current situation (have arrived and not arrivey or arrived)y (5) the current situation is present (have arrived and not had arrived).
If this expression is translated into Russian (Ya prishla)y then in the Russian sentence different elements of the situation will be realised linguistically: (1) the speaker (ya and not ti or mi) y (2) the speaker is feminine by gender \prishla and not prishol)y (3) she has arrived and not left (prishla and not vishla)y (4)] she walked and did not drive (prishla and not priehala)y (5) the event has aready happened and will not happen in the future (prishla and not pridot)y (6) the activity has been completed and is not in the process of being completed (prishla and not prihodila).
6. The concept of equivalence in the theory of translation
The English and the Russian sentences have only three of these elements in common: the speaker, the arrival, and the prior event, but the fact that these three elements are shared is enough to make the English and the Russian sentence mutu
ally interchangeable in the given context. Thus, while there is no formal correspon
dence between the English verb arrive and the Russian verb prishla, the English sentence I have arrived and the Russian sentence Ya prishla are interchangeable in the given situation and therefore they may be considered text equivalents. So, in Catford’s system the main criterion for text equivalence is the identity of con
textually relevant features.
6.4. Nida’s view on equivalence
Another classic of translation studies, Eugene Nida, in his book Toward a Science of Translating (1964), claims that it is not the identity of situationally relevant features that is the main criterion for equivalence, but rather the identity of the receiver’s reaction. He distinguishes between two types of equivalence: formal equivalence and dynamic equivalence.
He interprets formal equivalence completely differently from Catford. He only speaks about formal equivalence if the translator attributes priority to the source language text, and tries to render the source language text as faithfully as possible, not only in its content but also in its form including (1) grammatical units, (2) con
sistency in word usage, (3) meanings in terms of the source context. To faithfully give back the grammatical units, (1) verbs are translated into verbs, and nouns into nouns, (2) the boundaries of the sentences remain unchanged, (3) punctua
tion, paragraphing, etc. also stay the same.
When is such translation necessary? According to Nida, this is how Plato’s dia
logues have to be translated to understand the essence of Plato’s philosophical system and to be able to follow the development of his terminology. Formal equiva
lence is also important if the translation is done for linguists, who intend to con
trast the comparable units of two languages. Formal equivalence is thus not to be rejected, but is to be treated as an important form of translation in the case of cer
tain texts and certain audiences.
Dynamic equivalence is the exact opposite of formal equivalence; here the trans
lator concentrates on the target language receiver. When reading a dynamically equivalent translation a bilingual and bicultural person can justifiably say “That is just the way we would say it” (166). The dynamically equivalent translation is "the closest natural equivalent” of the source language text. To produce a "natural”
translation, the translator has to bear in mind three important factors: (1) the receptor language and culture as a whole, (2) the context of the particular mes
sage, and (3) the receptor-language audience (1964: 167).
Taking into account the receptor language and the receptor-language culture is the first criterion of a "natural” translation for Nida. He cites J. H. Frere’s thoughts:
"the language of translation ought, we think, to be a pure, impalpable and invisi
ble element, the medium of thought and feeling and nothing more; it ought never to attract attention to itself...” (Frere 1820: 481, in Nida 1964: 167).
According to Nida, a natural translation involves two principal areas of adapta
tion, namely, grammar and lexicon. Nida considers grammatical modifications
easier, especially because these are "dictated by the obligatory structures of the receptor language”. During translation, the translator is sometimes required to change the word order, replace a verb with a noun, or a noun with a pronoun, etc.
Adjusting the lexical structure of the source language to the semantic require
ments of the receptor language is a much more demanding task. From this aspect Nida considers three lexical levels: "(1) terms for which there are readily available parallels, e.g., river, tree, stone, knife, etc.; (2) terms which identify culturally differ
ent objects but with somewhat similar functions: e.g., book, (...) and (3) terms which identify cultural specialities: e.g., synagogue, homer, ephah, cherubim ...”
(1964: 167). In the case of this latter group, their "foreign associations” can rarely be avoided, and if the cultures in question are distant from each other, then it is almost impossible.
The second criterion of "natural” translation in Nida’s system is the considera
tion of the context of the given text. Nida’s concept of context is interesting. First he mentions the intonation and rhythm of sentences, and then discusses the issue of rendering the stylistic characteristics of the original text. How is it possible to avoid vulgarisms in a sophisticated text, or, on the contrary, how is it possible to avoid producing a complicated legal document out of an ordinary text when the translator tries to avoid ambiguities in the translation? Translators also have to think about "the standards of stylistic acceptability for various types of discourse”
which differ radically from language to language: "What is entirely appropriate in Spanish, for example, may turn out to be quite unacceptable "purple prose” in English, and the English prose we admire as dignified and effective often seems in Spanish to be colourless, insipid and flat” (1964: 169).
The third criterion of "natural” translation in Nida’s work is taking into con
sideration the receptor-language audience. "This appropriateness must be judged on the basis of the level of experience and the capacity for decoding, if one is to aim a real dynamic equivalence” (1964: 170). Translators have to make sure that the translation produces the same effect in the receptor-language readers as the original piece did in the source-language audience.
As we have seen, Nida bases the notion of dynamic equivalence on three crite
ria, but it is his third criterion that has gained wide currency in the literature on translation theory. Dynamically equivalent translation, according to Nida, is gen
erally identified with translation producing the same effect and the same reaction.
An oft-cited Biblical example is the Lamb of God, which has to be translated as Seal of God to an audience, such as the Eskimos, who do not know the animal lamb.
The requirement of identical reaction is, of course, easily questionable, and Nida was also aware of this. He himself calls our attention to the fact that we can
not always tell how the original audience reacted to a particular piece. "On the other hand, one is not always sure how original audience responded or were sup
posed to respond”(1964:170). Still, identical receiver reaction, as the criterion of equivalence, was a good starting point for further research. The more absurd it seems to require similar reader reaction in the translation of books written several hundred years ago or in the translation of literary works from completely different cultures, the less absurd it seems to require similar reader reaction in the case of translations of scientific or technical works written for a contemporary audience.
6. The concept of equivalence in the theory of translation
6.5. Other views on equivalence
The concept of equivalence (Äquivalenz) is a central category of German transla
tion research. Otto Kade (1968) enumerates four types of equivalences: total equivalence (totale Äquivalenz), where a source language unit has a permanent equivalent in the target language (e.g., in the case of terms, or institutional names), optional equivalence (fakultative Äquivalenz), where a given source language unit has several equivalents in the target language (e.g., in German: Spannung, in English: voltage, tension, suspense, stress, pressure), approximate equivalence (approximative Äquivalenz), where the meaning of a source language unit is divid
ed between two target language equivalents (e.g., German: Himmel, English: heav- en/sky), and finally, zero equivalence (Null-Äquivalenz), where the source lan
guage unit does not have a target-language equivalent (e.g., in the case of cultural words and realia).
Another well-known view in the theory of translation is Gert Jäger’s view on equivalence, which he put forward in his book Translation und Translationslinguistik, published in 1975. Jäger distinguishes between communicative and functional equivalence. He speaks about communicative equivalence when the “commu
nicative value" (kommunikativer Wert) of the original text does not change in translation, that is, the translation produces the same communicative effect as the original text. So far this is similar to Nida’s concept, but according to Jäger, com
municative effect is primarily a psychological concept, and therefore cannot be studied with the tools of linguistics. What can, however, be studied linguistically is the “functional value" (funktioneller Wert) of the text, the preservation of which is a precondition of functional equivalence.
How can the functional value of texts be grasped? By functional values Jäger refers to the sum of the functions of linguistic signs, the sum of their meanings.
More precisely, it is not exactly their sum, because he differentiates three types of meanings (semantic, syntactic, and pragmatic) borrowed from semiotics, and con
siders them only partly preservable in translation. Within syntactic meaning, he only considers preservable grammatical functions, which participate in the actual
isation of meaning, such as topic-comment structure. From pragmatic meaning, he only takes into consideration meanings that are realised linguistically, for exam
ple those that occur in the style of the text. Therefore, in his view, the functional value of the original text is a composite of semantic meaning, topic-comment struc
ture, and pragmatic meanings realised linguistically; and this is what has to be pre
served in translation to achieve functional equivalence. According to Jäger, such pre
cise delimitation of meanings to be preserved will save us from making empty statements as the one that claims that a precondition of equivalent translation is the preservation of "content", "sense”, or "meaning”.
At the end of his discussion, Jäger returns to the separation of communicative and functional equivalence. While communicative equivalence cannot be described with the tools of linguistics, functional equivalence can. While a functionally equiv
alent translation is equivalent communicatively as well, the reverse is not true, because the same original text can have several communicative equivalents, whose functional values are different (Jäger 1975: 87-95, 100-109).
Similarly to Jäger, in Yazik i perevod (1975) Barkhudarov also works with the three dimensional meaning concept of semiotics. He uses the terms (1) “refer- entsiaTnoe znachenie”, that is, referential meaning to indicate the relationship between signifier and signified, (2) “vnutriyazikovoe znachenie”, that is, intralin- guistic meaning to indicate the relationship between linguistic signs, and (3)
“pragmaticheskoe znachenie”, that is, pragmatic meaning to indicate the relation
“pragmaticheskoe znachenie”, that is, pragmatic meaning to indicate the relation