Approaches to the relation norm

According to Nida (1964: 136), any translation “must clearly reflect the meaning and intent of the source”. Nida (1964: 126–139) sees two basic, polar types of translation, between which there are other standards of translation representing the two basic orientations to a varying degree. At one end, formally equivalent translations (equivalence meaning correspondence, not identity, which Nida re-gards as impossible) emphasise the form and the content of the source text. At the other end are translations aiming at dynamic equivalence, which means that

“the relationship between receptor and message should be substantially the same as that which existed between the original receptors and the message”

(Nida 1964: 129). ‘Response’ appears to refer to the way the translated message and the original are understood by their respective audiences (Nida 1969: 95). A dynamically equivalent translation is defined as “the closest natural equivalent to the source-language message” (Nida 1964: 136). ‘Natural’ means that the translation adapts to the grammar and lexicon of the target language in order to be stylistically acceptable, fits into its new context and culture, maintains the spirit of the original, accurately reflects the emotional tone of the original au-thor, and takes into account the level of experience and understanding of the re-ceptor-language audience. The term ‘natural’ thus covers the language of the target text as well as some aspects of the relation between the source and the target texts. Nida (1964: 139) acknowledges the uncertainty in determining the way the readers of the source text “responded or were supposed to respond”.

Still, he argues that reader response is an important factor for all translating and therefore also for any evaluation of translation. Description of the different types of translation should be kept separate from the assessment of their appro-priateness for different types of audiences.

House (2001) regards equivalence as “a concept constitutive of translation”

(House 2001: 247). House’s notion of equivalence recognises the ambiguity of linguistic items and the fact that languages relate differently to reality.

5 See, for instance, Pym (1995), and Halverson’s (1997) analysis of how different views on the philosophy of science are behind the ways in which equivalence is regarded, on the one hand, in the linguistically oriented school and, on the other hand, in the historical-descriptive approach.

60

lence is a relative concept and related to the preservation of meaning on the se-mantic, pragmatic and textual level. Translation is defined as “the recontextuali-zation of a text in L1 by a semantically and pragmatically equivalent text in L2”

(ibid.). There are two main types of translation: an overt translation does not attempt to have the same function as the original but allows its readers to un-derstand the original text and its impact in its original culture. The aim of a cov-ert translation, however, is to have the same individual textual function and genre as the original. To achieve this, the translator applies a cultural filter, which means that the differences in conventions, rhetorical styles and expecta-tion norms have to be analysed and, when necessary, taken into account in translation at the levels of language/text and register to create a functionally equivalent translation (House 2001: 250–251). Furthermore, translations should be theoretically separated from versions which, as House remarks, are the results of client satisfaction and consumer service taking priority over the pursuit for equivalence.

Gutt (2000), in turn, excludes covert translations from the domain of trans-lation theory. In his account of transtrans-lation, which is based on the relevance the-ory developed by Sperber and Wilson (1986), equivalence of message or func-tion is not enough to distinguish translafunc-tion from other interlingual communi-cation. That distinction has to be made on the basis of the way the target text is intended to achieve relevance (i.e. to have large contextual effects with small processing effort needed to retrieve them) (Gutt 2000: 31 citing Sperber and Wilson 1986: 125), which is “in virtue of its resemblance with the other utter-ance” (the source text) (Gutt 2000: 210). A proper relation between a source and a target text would thus include the knowledge or implication that the target text is in fact a translation. Translation proper is interlingual interpretive use of lan-guage, and a translated utterance is presumed to interpretively resemble (i.e.

share explicatures and/or implicatures with) the original in a way which is con-sistent with the presumption of optimal relevance (pp. 46, 105–106).

Gutt sees covert translations as something that general translation theory does not need to concern itself with because:

- covert translations are instances of descriptive use of language (i.e. use of language in which “a mental representation or thought [is entertained] in virtue of its being true of some state of affairs” [Gutt 2000: 39]), intended to achieve relevance not as translations (i.e. as cases of interpretive use) but through descriptive accuracy, and therefore they can be written independently of the source text (p. 59)

- the client who commissions a covert translation is interested in the resulting text being an accurate description or an effective text, not in its interpretive resemblance with the original (p. 217)

- the reader does not care whether the text s/he is reading is a translation or not (ibid.), but, since s/he is not aware of reading a translation, s/he might not read it against cor-rect contextual knowledge (p. 222).

61

In such cases Gutt advises clients against translation and recommends pro-duction of original texts in the languages in question. A focal point in Gutt‘s (2000: 64) critique appears to be “a tendency to use the word ‘translation’

rather loosely to refer to almost any instance of communication that involves the transfer of information from one language to another”.

In other approaches, e.g. Pym (1995) and Koller (1995), equivalence is re-garded as something that defines translation and separates it from non-translation. According to Pym (2010: 6–7), equivalence is based on the idea that there can be “a relation of ‘equal value’ between a source-text segment and a target-text segment” which can be established on any linguistic level. There are two ways of thinking about equivalence: natural equivalence means that a translation reflects a correspondence which exists between languages and which can be verified by back-translation. Directional equivalence (Pym 2010: 25–30) assumes that a translator chooses between alternative translation strategies, and that translating a completed translation into its source language may not lead back to the point of departure. Pym (1995: 167) sees the translator as “an equivalence producer, a professional communicator working for people who pay to believe that, on whatever level is pertinent, A is equivalent to B”.

Other approaches are not so concerned about classifying different types of translation and defining translation. The focus shifts toward the factors that lead a translator to choosing a relation between a source text and its translation. In descriptive translation studies a text may be studied as a translation based only on the assumption that it is a translation (Toury 1995: 33–35). Equivalence is taken to mean “any relation which is found to have characterized translation under a specified set of circumstances” (Toury 1995: 61). This makes clients – the initiators, buyers, receivers and senders – key figures in defining equiva-lence.

Also the interaction between the translator and the client gets more atten-tion. Malmkjær (1993) refers to Davidson (1986) and bases her model of trans-lation on the meeting of the participants’ prior and passing theories in interac-tion through translainterac-tion. The translator’s knowledge of the requester’s aims is described as an important factor affecting the translator’s translational choices and thereby the relation which the translator creates between the source text and the target text. Malmkjær (1993: 146–147) mentions, first, “the requester-factor” and the requester’s aims for the translation together with the source text, and, second, the constraining influence of the source text on the target text as

“the two most powerful causes of tension in the process of performing a transla-tion” (Malmkjær 1993: 147). Other factors include the translator’s assumptions and knowledge of the writer of the source text, the reader of the translation, (in-tended) uses of the two texts and their contexts and the discourse surrounding them (Malmkjær 1993: 145).

Skopos theory assigns an important role to the client. The relation between the original and its translation (translatum) is the result of the translator’s deci-sions that s/he makes guided by the purpose, or skopos, of the translation as set

62

by the client together with the translator (Vermeer 1996: 7). If the client’s opin-ion about the way in which the source text should be translated differs from that of the translator’s, the translator should be able to negotiate with him/her in or-der to convince him/her of the superiority of his/her expert opinion (Vermeer 1996: 35).

According to Vermeer (1996), the author of the original and people who read and interpret his text, among them his translator, all interpret the world, includ-ing the text, differently. A translation does not have to be retrospectively

“equivalent” to, or automatically tied to, a particular source-text interpretation, only prospectively “adequate” to the target-text skopos. “Equivalence” is always partial (Vermeer 1996: 77–78). Equivalence of any kind, fidelity to the source text, imitation of the source text structure, etc. are legitimate and possible rela-tions which a translator may try to establish between the original and the trans-lation, if that is what the skopos expects him/her to do.

Although there is no pre-determined relation between a source text and its translation, “translation” is a specific type of translational action, namely, source-text based translational action. Depending on the relation between the source text and the target culture, or on the skopos, translation may not be pos-sible or purposeful, in which case the source text may have to be “rewritten”,

“paraphrased” or completely “re-edited”, requiring “translational action” other than translation (Vermeer 1989: 184–185).

Vermeer’s (1989) idea of quality is based on the skopos (purpose) of the translation. The translator’s task is not tied to equivalence of responses or to the preservation of a communicative intention. Vermeer writes that “[w]hat the skopos states is that one must translate, consciously and consistently, in accor-dance with some principle respecting the target text” (Vermeer 1989: 182), and further that “[a] translation must function in such a way that the given goal is attained” (Vermeer 1989: 186). The skopos must be defined separately for each translation commission in negotiations with the client or, if necessary, based solely on the translator’s expert judgement. Skopos theory appears to have a prescriptive undertone: on the one hand it states that translation is action and therefore automatically has an aim, but on the other hand, demands that the aim be consciously and systematically pursued. In translation quality assess-ment the question is then not whether a translation has a skopos but whether it is a suitable skopos and how well the translation fulfils it. Martín de Léon (2008: 14–15) observes, however, that the assessment of the purposes of trans-lation is not included in Vermeer’s theoretical thinking, but that Nord has tried to address that issue by her ethical principle of loyalty.

Also in Nord’s (2005: 32) functional approach to translation the intended or demanded function of the target text determines the relationship between the target text and its source text. The translation process begins when the initiator decides that s/he needs the target text for a certain purpose. The initiator may have more than one role in the process: s/he may also be the source text pro-ducer, the source text receiver, the target text receiver, the target text sender,

63

etc. (Nord 2005: 6–7). To serve the chosen purpose the target text must meet certain requirements, which the initiator defines in the translation assignment.

If the initiator cannot specify the requirements, the translator who is an expert in intercultural communication and the target culture, uses the information that s/he gets from the initiator about the target text situation and the purpose of the communication to set the skopos for the target text himself/herself (Nord 2005:

9–10).

The translator must also apply the loyalty principle in choosing a function for the target text, i.e. respect the wishes and expectations of the parties involved (for more on loyalty see section 6.2). The loyalty principle requires the transla-tor to consider the expectations and wishes of the initiatransla-tor, the target text re-ceiver and the original author who want a particular kind of relationship be-tween the original and the translation, or bebe-tween originals and translations in general. While skopos theory allows the translator considerable freedom as long as the skopos is served, “[l]oyalty limits the range of justifiable target-text func-tions for one particular source text and raises the need for a negotiation of the translation assignment between translators and their clients” (Nord 1997: 126).

Considerations regarding loyalty can thus be used to assess the suitability of a skopos for a particular situation.