Equivalence in translation

By Marouane Zakhir,
English translator,
University of Soultan Moulay Slimane, Morocco

prof.zakhir at gmail com

Become a member of TranslationDirectory.com at just $12 per month (paid per year)

Much ink has flown on discussing the term equivalence in translation. Nida (1964), Newmark (1981), Jacobson (1959-2000) and Bayar (2007) have written extensively on the nature, types, and degrees of equivalence in translation, whereas its opponents like Broek (1978), Mehrach (1997) and Leuven Zwart (1990) considered it an impossible point for the translator to reach and a hindering matter in the development of translation theory. The aim of this discussion is to shed light on writings which have dealt with equivalence in translation.

a.  Equivalence and contemporary equivalence theories

The increase in studying equivalence in translation coincides with the birth of a strong wave of research in machine translation. Leuven Zwart (1990:227 cited in Mehrach, 1997) stated:

It [equivalence] was used then in its strict scientific sense, to refer to an absolute symmetrical relationship between words of different languages. [1]

The aim of researchers to develop automatic translation led them to concentrate on the equivalent effects that exist between words from different languages. This gave an impetus to research in equivalence in translation.

The Russian-born American structuralist Roman Jacobson (1959-2000: 114) is one of the earliest theorists who were occupied by the study of equivalence in meaning. Jacobson claimed that "there is ordinarily no full equivalence between code units" (cited in Munday, 2001). [2] To confirm his idea, Jacobson uses the example of ‘cheese’ which does not have the same equivalent of the Russian term ‘syr.’ He stated that Russian does not have the concept ‘cottage cheese’ in its dictionary and suggested translating it by ‘tvarok’ instead. Jakobson also pointed out that the problem of both meaning and equivalence is related to the differences between structures, terminology, grammar and lexical forms of languages. He stated that "e quivalence in difference is the cardinal problem of language and the pivotal concern of linguistics." (cited in Munday, 2001) [3]

In his work on Bible translation, Nida (1964) devotes much of his research to meaning in both its semantic and pragmatic natures. He rejects sayings which regard meanings of words as fixed and unchanged, and suggests giving meaning a more functional nature. For him, words get their meanings through the context and culture in which they are used. Nida also distinguishes between many types of meaning: linguistic meaning, referential meaning and emotive meaning (Munday, 2001). [4]

It should be noted that Nida’s concept of meaning in translation is influenced by the Chomskyan theory of ’generative transformational grammar.’ Chomsky believes that each language is composed of a deep structure that undergoes the process of transformations and a surface structure produced by these transformations and is subject to phonological and morphophonemic rules. In his translation of the Bible, Nida adopts this theory and gives much concern to the deep structure, which contains the core of meaning. Yet, Nida’s treatment of meaning is different from that of Chomsky. Edwin Gentzler (1993) [5] said that:

Chomsky investigates the meaning inherent in the sign cut off from cultural context; Nida’s primary concern is not with the meaning any sign carries with it, but with how the sign functions in any given society.

The negligence of cultural context in the generative theory is the main difference between Chomsky’s and Nida’s study of meaning. However, they both share the same view on the nature of languages as including deep and surface structures (E. Gentzler, 1993). [6]

Nida’s theory of translation is characterized by his distinction between two types of equivalence: formal equivalence and dynamic equivalence. In formal equivalence, the translator focuses on the message itself, that is, its form and content, as there should be a close similarity between the ST and the TT message (Nida, 1964). [7] This source-oriented type is described by Kelly (1979: 131 cited in Mehrach, 1977) [8] as an approach that "depends on one-to-one matching of small segments, on the assumption that the centre of gravity of text and translation lies in the significance for terminological and artistic reasons."

In the same context, Munday, (2001) [9] pointed out that ‘gloss translation’ with scholarly ‘footnotes’ are the most typical of formal equivalence, for they help the reader to understand the source culture’s language and customs.

Concerning dynamic equivalence, Nida mentioned that this type is based on "the principle of equivalent effect," in which "the relationship between receptor and message should be substantially the same as that which existed between the original receptor and the message." (Nida, 1964: 159, cited in Munday) [10]

Nida gives paramount importance to the notion of ‘naturalness.’ He argues that the main aim of ‘equivalent effect’ is to achieve "the closest natural equivalent to the source language" (Nida, 1964). [11] Actually, ‘naturalness’ as a basic keyword in Nida’s theory relies on the adaptation of grammar, cultural references and lexicon of the ST. It goes without saying that Nida privileges the preservation of the text meaning on its style, since it allows the translator to create the same equivalent effects.

In short, Nida aims in his book Towards A Science of Translation is to redefine the principles which evaluate the sufficiency of translation (Gentzler, 1993). [12] Comparing form and content of texts, Nida mentions that content should come first in translation. He argues that formal translators who focus more on forms of poetry, for instance, are more likely to misinterpret the "intention of the author" and more apt to "distort the meaning" (Nida, 1964). [13] According to Nida, the dynamic translator is more faithful than the literal one, since he (DT) may preserve "more fully and satisfactorily the meaning of the original text" (Nida, 1964). [14] Also, Nida’s notion of ‘equivalent response’ is important for translators to achieve a successful translation (Munday, 2001). [15]

It should also be noted that Newmark’s distinction between ’communicative translation’ and ’semantic translation’ in his book Approaches to Translation (1981) [16] is similar to Nida’s types of equivalence. For ’communicative translation’, which tends to create the same effects on the readers of the TT as those obtained by readers of the ST, resembles Nida’s notion of dynamic equivalence, whereas ’semantic translation,’ which focuses on rendering the contextual meaning of the ST according to the syntactic and the semantic characteristics of the TT, is similar to Nida’s formal equivalence.

However, many critics of the ’equivalent effect’ by Newmark appeared in his Textbook of Translation (1988). Newmark sees Nida’s ’equivalent effect’ as:

The desirable result, rather than the aim of any translation . […] It is an unlikely result in two cases: (a) if the purpose of the SL text is to affect and the TL translation is to inform (or vice versa); (b) if there is a pronounced cultural gap between the SL and the TL texts. [17]

We infer from this quotation that the ’equivalent effect’ is a result which all translators long to achieve. However, this result can be unachievable if the SLT and the TLT do not share the same goal; i.e., to inform or to affect, or if they do not have the same cultural equivalents. The possession of cultural references, together with the remoteness in time and space, reduces the possibility of achieving ’equivalent effects,’ except in case the reader is imaginative, sensitive and has a good knowledge of the SL culture (Newmark, 1988). [18]

Furthermore, Newmark (1988) argues that the text may reach a ’broad equivalent effect’ only if it is ’universal,’ as in this case the ideals of the original text exceed all cultural frontiers. [19]

Another figure of translation theorists who devoted a great deal of research to the notion of equivalence is Koller (1979). He distinguishes between five types of equivalence: ’denotative equivalence’ refers to the case where the ST and the TT have the same denotations, that is, conveying the same extra linguistic facts; ’connotative equivalence,’ also referred to as ’stylistic equivalence,’ is related to the lexical choices between near synonyms; ’text normative’ refers to text types, i.e., the description and analysis of a variety of texts behaving differently; ’pragmatic equivalence,’ also called ’communicative equivalence,’ is oriented towards the receptor of the text, who should receive the same effect that the original text produces on its readers; ’formal equivalence,’ may also be referred to as ’expressive equivalence,’ is related to the word-for-word rendering of forms, aesthetic and stylistic features of the ST.

Koller (1979: 176-91, cited in Munday, 2001) [20] also dedicated a remarkable part of his research to the examination of the relation between ‘equivalence’ and ‘correspondence.’ The former examines the equivalent items in both the ST and the TT, and is based on De Saussure’s parameter of ‘langue,’ while the latter can be related to contrastive analysis, and is based on the De Saussure’s ‘parole.’

Moreover, the term equivalence continued to be a crucial point of debate for many years. Theorists tried to define it as a way to enhance its role in translation. According to Broek (1978), Catford stated:

Translation equivalence occurs when an SL [source language] and TL [target language] texts or items are related to (at least some of) the same relevant features of situation substance. [21]

Similarly, in her study of equivalent types in translation, Baker argues that equivalence is always relative in the sense that it is influenced by many linguistic and cultural factors (Baker, 1992). [22]

The development in equivalence research is also characterized by the work of Bayar (2007). In her book To Mean Or Not To Mean , Bayar distinguishes between formal equivalence, semantic equivalence, cultural equivalence and pragmatic equivalence. For her, formal equivalence "designates an area of correspondence ranging around the word, albeit involving lower units such as the phoneme or the morpheme." [23] She argues that transliteration, categorical correspondence such as the correspondence of ’noun to noun and verb to verb’ between ST and TT, and textual correspondence such as length, stylistic aspects, meter, rhythm and rhyme are all instances of ’formal equivalence’ (Bayar 2007).

As far as semantic equivalence is concerned, Bayar (2007) [24] notes that this type relies on the preservation of many semantic criteria: denotation, connotation and propositional content. According to her, words which do not have the same equivalent meanings could be translated by ’explanatory expressions’ as a way of compensation. For instance, the English word ‘nod’ that has no equivalent word in Arabic, can be translated by the expression /?anζama bi ra?sihi/ (p.163-7). For the third type, ’cultural equivalence’, Bayar (2007) considers it to be the most difficult and ’controversial kind of equivalence’ due to it is relation with ’human identity.’ She defined it as follows:

Cultural equivalence aims at the reproduction of whatever cultural features the ST holds into the TT. These vary from things specific to the geographical situation, the climate, the history, the tradition, the religion, the interpersonal or inter-community social behavior, to any cultural event having an effect on the language community . [25]

It is clear from this definition that ’cultural equivalence’ consists of the rendering of the SL cultural features into a TL in a way that helps the reader understand these foreign cultural aspects through his own cultural ones. Actually, ’cultural equivalence’ can be easily reached when the cultural words under translation are universally known. However, this can be diminished with cultural differences that languages may show. Further, Bayar (2007) [26] discusses the importance of preserving the author’s ideology to make the translation equal to the ST.

As far as ’pragmatic equivalence’ is concerned, Bayar (2007) [27] argues that this type tends to reproduce the context and text goals of the SL. She also shares the same idea with Hatim and Mason (1990: 236-8) that "pragmatic equivalence subsumes all of the semio-pragmatic-communicative layers of communication." [28] Examples of these semiotic and communicative dimensions are genre, field, mode, tenor, text type and translation purpose (skopos).

In brief, it is true that Bayar’s types of equivalence have already been tackled by western theorists, but her illustration of equivalence enhances its importance in translation studies.

However, the notion of equivalence or equivalent effect is not tolerated by all theorists. Many scholars reject its existence in translation. In his essay The Concept of Equivalence in Translation , Broek stated, "we must by all means reject the idea that the equivalence relation applies to translation." (Broek, 1978) [29]

Broek refuses the idea of equivalence in translation as a form of linguistic synonymy. He argues that synonymy does not exist even with words of the same language (p.34). Besides, Broek rejects terms like similarity, analogy, adequacy, invariance and congruence, and the implications they may have in translation. He also redefines the term equivalence by the concept of "true understanding" (p.29).

In the same context, Leuven-Zwart notes that the concept of equivalence "not only distorts the basic problem of translation, but also obstructs the development of a descriptive theory of translation" (Leuven Zwart, 1990: 228 cited in Mehrach). [30] Leuven Zwart also mentions that equivalence proponents relegate the importance of crucial factors such as ’the situation of the utterance,’ ’the intention of the speaker’ and ’the effect on the hearer.’

Similarly, the Moroccan scholar Mehrach (1997) considers equivalence "an impossible aim in translation." He claims that no two languages share the same linguistic structures and social or cultural aspects. He rather proposes the use of the term ’adequacy’ as a reference for the ’appropriate’ translation, that is, "a translation that has achieved the required optimal level of interlanguage communication under certain given conditions." [31]

 In brief, it is clear from the above opposing views that the notion of equivalence is controversial and relative in nature. It is, in fact, difficult to determine since no one could objectively define the point at which the TT becomes equal to the ST. Perhaps, equivalence should be considered as a form of approximation in which the TT approximates the ST. It can also be rated on a scale that ranges from optimum to zero degree (Bayar, 2007) [32].

b. Degrees of equivalence

Equivalence consists of seven degrees: optimum translation, near-optimum translation, partial translation, weaker and stronger translation, poor translation, mistranslation and zero equivalence/non-translation. Each degree has specific characteristics that differentiate it from the other. To distinguish these degrees, much emphasis will be put on pragmatic and cultural aspects as measures to assess the degree of preservation of the ’superordinate goal’ of the ST.

Optimum translation

It refers to the highest level of approximation to the ST. Bayar (2007) defines it as "the closest equivalence degree attainable, given the circumstances, the linguistic and extralinguistic resources actually available to the translator." [33] In other words, a TT may reach the optimum degree when it preserves the ’superordinate goal’ of the ST and its five requirements (i.e., genre, field, mode, tenor and type). The TT should be semantically and grammatically well-formed, with sentences that cohere to each other to serve the ST goal and to preserve its content. Any deviation from these characteristics distances the translated text from the optimum degree. The following examples provide an illustration of the optimum degree of translation:

a)

 1-Eng ST: He was armed to his teeth.

 2-Arb TT1: كان مسلحا حتى أسنانه   

 3-Arb TT2 , (optimum): كان مدججا بالسلاح 

b)

 1-Eng ST: He kicked the bucket.

 2-Arb TT1: ركل الدلو 

 3-Arb TT2, (optimum): وافته المنية 

Despite their readability and grammatical accuracy, the examples (2) of these idiomatic expressions do not reach the optimum degree of translation. This is due to their distortion of the ST’s meaning. On the contrary, examples (3) are optimum because they succeed in carrying the same implicatures and cultural aspects of the STs. In brief, optimum translation is feasible. It is easy to achieve when the ST is simple and does not contain cultural aspects. To clarify this, see the example below.

c)

 1-Eng ST: Zidan shoots the ball.

  2-Arb TT1: Zidan frappe le ballon. (Optimum)

 3-Arb TT2 : يضرب زيدان الكرة   (Optimum)

It is obvious from the above example that the simplicity of the ST helps to reach the optimum degree of translation. Yet, problems with optimum translation rise when dealing with literary texts and more specifically poetry, where translation is governed by many aesthetic and stylistic rules.

Near-optimum translation

Near-optimum translation refers to the case where the ST superordinate goal and sub-goals are cohesively and coherently rendered to the TT but do not reach the readability of the optimum degree from a textual point of view. For the sake of clarification, we will use the example given by Bayar (2007). [34]

d)

SL: If you happen to have read another book about Christopher Robin, you may remember that he once had a swan.

TT1: S’il vous est arrivé de lire un autre livre sur Christopher Robin, vous pourriez peut-être vous rappeler qu’il avait un cygne. (Near optimum)

TT2: S’il vous est arrivé de lire un autre livre qui parle de Christopher Robin, vous vous rappelleriez alors qu’il avait un cygne. (Optimum)

Reading this example, we notice that the French version TT1 wrongly uses the adverb ‘sur’ and the verb ‘peut-être’ in translation. This negatively affects the smooth readability of the TT. Conversely, the TT2 is an example of optimum translation because it preserves the smoothness of its readability.

Partial translation

This type refers to the partial rendering of the ST’s superordinate goal to the TT. It should be noted here that the readability and accuracy of the TT do not mean its preservation of the ST, for the TT might be read smoothly, without conveying the ST goal.

e)

Eng, ST: Never too old to learn.

Arb, TT1: (partial translation) ليس للتعلم سن يحده 

Arb, TT2: (optimum translation)  أطلبوا العلم من المهد الى اللحد

Here, the TT1 does not convey the whole superordinate goal of the ST. Therefore, it is considered as partial translation while TT2 is an optimum one.

Weaker and stronger versions

Using Bayar’s words, some translations are called weaker versions because they reproduce the ST goals in ’attenuated terms’ if compared to the original, whereas others are named strong versions for their use of stronger terms in their rendering of the ST goals. [35] To clarify these types, let us observe the differences in the following examples:

f)

Eng, ST: Once bitten, twice shy.

Arb, TT1: (weaker version) عندما تلدغ مرة تصبح خجولا مرتين 

Arb, TT2: (optimum) لا يلدغ المؤمن من جحر مرتين 

Arb , TT3: (stronger version) كثرة الخجل تأتي من اللدغ 

The distance or approximation of these versions (weaker/stronger) from the optimum translation depends on the degree of their alteration of the ST goal.

Poor translation

In poor translation, readability is the core of the problem. Though the TT may preserve the ST superordinate goal, it is read with great difficulty by the receptor. In other words, poor translation occurs when the translator fails to transfer the ST goal into a readable TT. Below is an example of poor translation.

g)

Arb, ST: يلومونني في حب ليلى عواذلي ولكنني من حبها عميد 

Eng, TT1: (poor translation). They blame me for loving Laila / but I am with her love smitten.

TT2 (optimum): They blame me for loving Laila / but I am deeply smitten with love for her. [36]

The TT1 shows a poor translation because the reader cannot easily comprehend the ST goal.

Mistranslation

In mistranslation the TT neither sounds readable nor preserves the superordinate goal of the ST.

i)

ST: It is raining cats and dogs.

TT1: (mistranslation)  انها تمطر قططا وكلابا

TT2: (optimum)  ينهمر المطر مدرارا

Here, we see that the TT1 does not only distort the superordinate goal of the ST, but it is also out of context.

Zero equivalence

Zero equivalence occurs when there is no one-to-one equivalent between the ST and the TT. This happens when the translator deals with texts which contain culturally-bound words or expressions. Examples of these words are: ‘kassāl’ and ‘innur’ in Moroccan Arabic and ‘nuts’ in English (see Bayar, 2007). [37] In fact, zero equivalence rarely occurs at the text level, except in some literary forms as poetry and fairytales. The translator in this case may resort to translation recreation.

All in all, equivalence in translation can be measured by a scale that ranges from optimum equivalence to zero equivalence. These degrees of equivalence might be measured by the levels of approximation or distance from the ST’s ‘superordinate goal.’ While optimum equivalence is considered as the highest level in translation, or the most approximate degree from the ST, zero equivalence is the lowest degree or the most distant degree from the ST goal.

[1] Mohamed Mehrach. (1977) Towards a Text-Based Model for Translation Evaluation . Ridderkerk: Ridden print, p. 14.

[2] Jeremy Munday. (2001). Introducing Translation Studies, Theories and applications . London and New York: Routledge, p. 36.

[3] Ibid, p. 37.

[4] Ibid, p. 38.

[5] Edwin Gentzler. (1993). Contemporary Translation Theories, London and New York: Routledge, p. 53.

[6] Ibid., p. 55.

[7] Eugene A. Nida, (1964). Toward a Science of Translating . Leiden: Brill, p. 159.

[8] Mohamed Mehrach, op. cit ., (1977), p. 44

[9] Jeremy Munday, op. cit ., (2001), p.41.

[10] Ibid, p. 42.

[11] Eugene A. Nida, op. cit ., (1964), p. 166.

[12] Edwin Gentzler. (1993). Contemporary Translation Theories, London and New York: Routledge, p. 58.

[13] Eugene A. Nida, op. cit ., (1964), pp. 191-2.

[14] Ibid, p. 192.

[15] Jeremy Munday, op. cit ., (2001), p. 42.

[16] Peter Newmark, (1981). Approaches to Translation . Oxford and New York: Pergamon, p. 39.

[17] Peter Newmark. (1988). A Textbook of Translation . London and New York: Prentice Hall International (UK) Ltd, p. 49.

[18] Ibid, p. 48.

[19] Ibid, p. 49.

[20] Jeremy Munday, op. cit ., (2001), pp. 46-7.

[21] Broek, Raymond Van Der, (1981). "The Limits of Translatability Exemplified by Metaphor Translation", in Itamar Even-Zohar and Gideon Toury (eds) Translation Theory and Intercultural Relations, Poetics Today , p. 38.

[22] Mouna Baker. (1997). The Routledge Encyclopedia of Translation Studies, Part II: History and Traditions . London and New York: Rutledge, p. 6.

[23] Monia Bayar, (2007). To Mean or Not to Mean , Kadmous cultural foundation. Khatawat for publishing and distribution. Damascus, Syria, p. 163.

[24] Ibid, pp. 163-7.

[25] Ibid, p. 177.

[26] Ibid, pp. 186-203.

[27] Ibid, p. 206.

[28] Ibid, p. 208.

[29] Broek, Raymond Van den, (1978) "The Concept of Equivalence In Translation Theory: Some Critical Reflections", in J. S. Holmes, J. Lambert and R, Van den Broek (eds), Literature and Translation , Leuven: Academic, p. 33.

[30] Mohamed Mehrach, op. cit ., (1977), pp. 14-15.

[31] Ibid, p. 16.

[32] Monia Bayar, op. cit ., (2007), pp. 213-223.

[33] Ibid., p. 214.

[34] Ibid, p. 220.

[35] Ibid., p. 221.

[36] Mohammed Addidaoui, op. cit ., (2000), p. 32.

Published - April 2009

Submit your article!

Read more articles - free!

Read sense of life articles!

E-mail this article to your colleague!

Need more translation jobs? Click here!

Translation agencies are welcome to register here - Free!

Freelance translators are welcome to register here - Free!