Intuition as to relevance is generally taken as given; individuals are assumed to be able to make distinctions between information that is relevant and information that is irrelevant. The problem is in proving this intuitive function. In language, ideas of relevance are vague and subjective, with no clear or concrete meaning; this makes understanding the phenomenon difficult (Sperber & Wilson 1995 119).
Alves and Gon�alves (2010) suggest that, when discussed as a function of �intuition,� relevance is not a clear concept. Delving further into the field, however, it becomes clear that there are numerous ways in which the notion of relevance can be understood. There are two definitions that are useful in this regard. The first is the Gricean definition of relevance, which looks specifically at relational relevance and assumes that speakers indicate directly when the information they are communicating is relevant and when it is not (Grice 1975). The second definition is called the relevance-theoretic this is a theory of relevance that combines ideas from both a linguistic-pragmatic and cognitive approach. It is designed to discern relevance through the process of inference.
Gutt (1991) proposes a relevance theory framework that adapts the explanatory approach and can provide an account of translation decisions (Sang Zhonggang, 2006). In the relevance theory, communication is analysed from a competence perspective, rather than being treated simply as behaviour. The theory explains our ability to communicate with each other in terms of our mind�s information-processing faculties. The implications of this on how communication is interpreted can be applied to translation and can facilitate translators� understanding of the process of translation (Gutt, 2000).
By applying this theory, I shall analyse and discuss a text and its translation, as this theory helps to account for the decisions made by the translator (Sang Zhonggang, 2006). Thus, I shall start by analysing and comparing the contextual factors of a source text and its translation. I shall investigate how the context was reflected at the linguistic level and how this has affected the translation strategies used by the translator.
2. Methodology
The source text (ST) (See the appendix for ST and its translation) in question is a short story that is available online at World Association of Arab Translators and Linguists. This story was written by the Arab writer Muhammad Khaled Ramadhan and translated by Yousef Salah Al-Hajjar. This text and its translation, which is available on the same website, will be analysed by following the relevance theory (Gutt, 2000). Based on this theory, a top-down method of analysis will be utilised. Therefore, I shall start by analysing the context. Subsequently, I shall investigate how that context is reflected in the semantic representation, syntactic proprieties, semantic constraints, and formulaic expressions. For the purpose of this article, it was necessary to literally translate the examples used from the Arabic source text into English.
3. Relevance theory
It would be beneficial to introduce relevance theory before using it to discuss the decisions taken by the translator while analysing the translation of the text. Relevance theory refers to a theory that differentiates between the two different ways in which the human mind processes information, in terms of language mode the descriptive mode and the interpretive mode. Language is used descriptively �when it is intended to be taken as true of a state of affairs in some possible world.� On the other hand, an utterance is �used interpretively when it is intended to represent what someone said or thought� (Gutt 1998 44). Interpretation, according to relevance theory, occurs when one text or utterance is claimed as a representation of an original text, via a secondary voice. Thus, a translation can be understood as having an interpretive function; as such, it can be seen as an �interpretive use of language across language boundaries� (p. 44). There are several insights that have been particularly valuable with regard to the relevance theory. First, the theory is based on the assumption that interpreting a message requires a complex inferential process that involves a combination of linguistic coding, face-value semantics, and context-based inference. Second, message recipients (readers and listeners) seek to optimise payoff with respect to processing cost, such that when the minimum cost of processing is exceeded, recipients seek increased payoff. Third, relevance theory also notes that there are cases, such as in figurative or poetic language, where message producers (writers and speakers) intentionally wish to convey an array of possible interpretations, rather than a single interpretation (Dolley, 2007). Finally, relevance theory recognises that making implicit information explicit in the text itself may not be the best way to address contextual discrepancies in translation (ibid).
Furthermore, relevance theory suggests that recognising a text as a translation and identifying the contextual discrepancies arising from differences between the source language's culture and the reader's culture can serve as a preliminary step in accessing the text's meaning, with respect to the reader's understanding of the world.
4. Text Analysis
4.1. Context
As explained by Gutt (2000), context does not represent external physical factors. However, it does represent the hearer�s mental environment where she processes an utterance. The potential shared context between the writer and the reader is considered to be an important factor in achieving successful communication. This success occurs when the reader�s expectations and writer�s intention meet, as has been highlighted by Gutt (1996).
As the target readers� expectations play an important role in the relevance theory, a consideration of the audience of the source and target text (TT) used in this article is of relevance. The ST utilises the short story genre; this is reflected, for example, in the use of temporal cohesive devices, such as baad thalek �after that�, thuma �then�, and baad heen �after a while�. The language is a mixture of standard and dialectic Arabic (e.g. thahro �his back�) (e.g. Jeet �I came�) and (e.g. khalethon �left them�). It can be said that there is an intimate relationship between the reader and the writer. This is also apparent in the use of cultural proverbs, such as ingata men thahro sita wa siteen kharzah �he lost sixty six vertebrae�, which means, �he was fascinated by her beauty�.
This story was written for Arab readers, the majority of whom are Muslim. Although the translation of this story sounds equally formal (although the register of the ST and the TL is not very high), the relationship between the writer and the reader appears to be detached. This is evident in the absence of any dialectical cultural proverbs or references in the TT. The choice of language (English) reflects the intention of the initiator (the person who wanted the story to be translated) to address a wider audience (English-speaking readers). Thus, it is evident that there is a discrepancy in the cultural and cognitive environments of the ST and the TT audiences and, as such, in their cultural and linguistic expectations.
��The central claim of relevance theory is that human communication crucially creates an expectation of optimal relevance, that is, an expectation on the part of the hearer that his attempt at interpretation will yield adequate contextual effects at minimal processing cost�� Gutt (2000 31). Zhonggang (2006) points out that the greater the contextual effects achieved by the reader, the greater the relevance of the text, and vice versa. These contextual effects result from the interaction between the new information and the context, either by strengthening an existing assumption, by contradicting and eliminating an existing assumption, by weakening an existing assumption, or by combining with an existing assumption to yield a contextual implication (Gutt, 1996). Hopefully, our analysis of the text will demonstrate how the translator attempts to compensate for the discrepancy between the ST and TT audiences� expectations, reducing the TR�s possessing efforts in understanding the text and increasing its degree of relevance to them and, thus the contextual effects that they can achieve. The level of contextual effects achieved by the target audience differs according to the degree of relevance (Zhonggang, 2006).
4.2. Stylistic features
From the point of view of relevance theory, we should be concerned not only with the information content of a certain utterance, but also with the way in which it is expressed in the text (Gutt, 2000). Therefore, we will examine certain stylistic features in the ST, such as quotations and the organisation of the text, and analyse how they are conveyed in the TT.
The ST utilised a large number of direct quotations in the conversation between the characters in the story. The translator of the text preserved most of these direct quotations in the TT.
For example
However, some of these direct quotations are transformed into indirect quotations by the translator. For instance
It is evident from the above examples that, in translating quotations, the translator sometimes depends on resemblance in linguistic proprieties (example 1). The translator in this example used the same syntactic structures, semantic representation, lexis, and other linguistic features. He depends on resemblance in cognitive effects only three times, for example, in example 2. There is a resemblance between 2 (a) and 2 (b) in terms of propositional content and implications, although they differ in linguistic features. In addition, by comparing the organisation of the ST and the TT, it becomes clear that the TT is divided into paragraphs, in contrast to the ST. The decisions taken by the translator concerning the stylistic features of the text can potentially be interpreted as an attempt to reduce the English-speaking reader�s processing effort and increase the degree of relevance to the text and, consequently, the contextual effects they may obtain.
4.3. Semantic representation
According to relevance theory, the interaction between the stimulus and the cognitive environment contributes to the communication of the meaning of the text (Gutt, 2000). Let us consider, for example, the conversation between the woodcutter in the story and his pregnant wife, as outlined in example 3.
In the ST, the man promised to give his wife (ngoot) when she gave a birth to a baby boy. In the Arabic culture, the word (ngoot) represents what is given to a child when she is born, which is, strictly speaking, gold or money. The translator in this example translated this word (ngoot) as �birth present�, which does not mean exclusively gold or money, as it is normally the case in the Arab culture.
Sperber and Wilson (198686) point out that the concept of relevance theory is associated with three entries or sets of information in our mind
��The logical entry for a concept consists of a set of deductive rules which apply to logical forms of which that concept is a constituent. The encyclopaedic entry contains information about the extension andor denotation of the concept that is, about the objects, events andor properties which instantiate it. The lexical entry contains information about the natural-language counterpart of the concept the word or phrase of natural language which expresses it��.
The logical and the encyclopaedic entries play an important role in the interpretation of the utterance. It is possible to say that the translator used the logical entry of the concept (ngoot), by translating it as a �birth present�. The word (ngoot) and its translation may not be semantically different, but in the translation it has lost its encyclopaedic entry. The translator may have aimed to reduce the TT audiences� processing efforts by simplifying the concept, because the meaning of this word may not be part of the English-speaking readers� cognitive environment. The interpretation of the word (ngoot) by the translator as �birth present� may not be sufficient for the TT readers in the same way as it is for the ST audience, because the TT readers may have not seen or heard of this word. However, the translator may have intended to make the text more relevant to the target audience, who rarely give gold or money as a birth present.
4.4. Formulaic expressions
Formulaic expressions, like the opening of a short story, do not enter the semantic representation and do not have a logical entry. However, their meanings are included in their encyclopaedic entries (i.e. what people know about them). This encyclopaedic information facilitates the interpretation of these formulaic expressions. Typically, expressions containing similar information are used to translate these formulaic expressions, as can be seen in the opening of this short story
The opening in the Arabic text contains Islamic and cultural concepts, such as Salo al Alnabi (pray for Muhammad) and Yakeram (Generous people). This opening used to be employed by storytellers when reading a story for a group of people in the Arabic culture. The translator sacrificed this encyclopaedic information and used an opening typically used in the target language (TL), making it more relevant to the English speaking target readers� expectations.
The semantic adjective based meaning of a proverb, on the other hand, seems to be clear; however, the general meaning of the proverb is usually difficult to grasp.
For example, the proverb Ingataa men thahri sit wa seteen Kharzh (broke of my back sixty six vertebrae), is in the original text.
It is easy to understand each of these words separately, such as (back), (broke), and (vertebrae), but it is difficult to understand the meaning of the proverb as a whole. In translation, translators tend to translate proverbs holistically, based on their encyclopaedic entries, rather than using their semantic meaning. In the translation of this short story, the translator prefers not to translate any of the proverbs in the ST. Typically; speakers of a foreign language find it difficult to understand proverbs when they hear them, which could lead the translator to delete them in the TT, thus reducing the TT readers� processing efforts.
4.5 Syntactic properties
The differences between languages can prevent the translation from preserving relevant syntactic properties in the ST, which may result in the ST losing part of its meaning in the TT (ibid). This can be demonstrated by the following example
In this example, the translator translated several sentences from the ST into only one sentence in the TT.
One of the factors that affect processing effort is ��frequency of use the more commonly a certain type of structure occurs in stimuli, the less processing effort it seems to require�� (Gutt, 2000 148). As a result, the ST writer depends on a style rich with repetition, which is typically used in Arabic writing.
Further, as the ��structure complexity of the stimulus�� affects the processing efforts, the translator chose to use a style that English readers will find easy to follow, reducing the repetition.
4.6. Interpretive resemblances between the thoughts and utterances
Each utterance contextually implies a number of assumptions. The�analytic implications which the communicator intends to convey�� are called explicatures, while ��the contextual assumptions which he intends to convey�� are called implicatures (Gutt, 2000 40). For example, when the writer of the ST mentions pilgrimage, he does not mention the place where the merchant is going to perform that pilgrimage (see example 6). As the writer is writing for Arab readers, the majority of whom are Muslim, in accordance with the principle of relevance, the audience�s first interpretation of this sentence will be consistent with the intention of the writer, that the merchant is going to Mecca to perform a pilgrimage. On the other hand, in the TT, the word �Mecca� has been added. This word will guide the TT reader in identifying the intended referent in hisher memory. The translator, in this case, facilitates the reduction of the processing effort for the target English speaking reader, who was not provided with the information that the merchant is Muslim and that, as such, he should go on pilgrimage to Mecca.
Some information must be made explicit in the TT, as the ST writer shares information with his readers that the TT readers do not have access to. The translator attempts to convey the set of assumptions contained in the ST sentence, including the explicatures and implicatures, in order to further explicate the TT sentence.
4.7. Idiomatic Translation
Advocates of relevance theory consider that providing the TT readers with adequate contextual effects is crucial for successful communication and for conveying the intended interpretation (ibid). However, this does not apply fully to the following example, which expresses the woodcutter�s feeling when he realises that he has a baby boy.
In this example, if the Arabic expression had been translated into English literally, it would probably have been understood by English speakers, who would obtain the required contextual effects. However, the translator chose a typical expression in English. This could be because the translator was attempting to reduce the target readers� processing efforts, increasing the degree of relevance for the target reader.
5. Discussion
Relevance is a ��comparative notion�� (Gutt, 2000 31) and the degree of relevance for each utterance differs in every context. Therefore, it might be difficult to evaluate the degree of relevance of the entire text in this article; however, it may be possible to say that the translator aimed to achieve optimal relevance. The examples provided above demonstrate that the translator tried to help the target audience to infer the intended interpretation of the original story. As the purpose of the communicator is to facilitate the understanding of his intention, it is his responsibility to express himselfherself so that the readers�listeners� mind will realise the intended meaning on the first interpretation, finding the text optimally relevant to himher (Sperber and Wilson, 1986).
As demonstrated in the text analysis, the translator is aware of the assumptions that exist in the TT readers� cognitive environment, which they depend upon to comprehend the text. Thus, the translator changed, explicated, and deleted several contextual codes in the ST, as shown in the above examples.
The translator made a few changes in the translation, in order to convey the same interpretation of the story, as this interpretation is affected by considerations of relevance, which is context-dependant. By transferring the assumptions associated with the ST (explicatures and implicatures) and by explicating implied information, he reduced the processing load placed on the target reader.
As the ST writer and readers share the same religious beliefs and cultural background, the writer uses many Islamic and dialectical proverbs. As the translation was intended for a wider audience, everybody who speaks English and is interested in reading such stories, the translator manipulated the Islamic and cultural references in the original story in a way that is not intended to be directed at a specific audience. An English speaking reader with little interest in Islamic or Arabic culture would be unable to find an interpretation consistent with the principle of relevance. She would not be sure what the ST writer was intending to say. However, these changes have provided the TT reader with adequate contextual effects, which make the text relevant to himher; as such, she will be satisfied. Despite the changes made by the translator in the TT, it is probably fair to say that he did not add any new information to the text, nor did he distort the general meaning of the original text.
We can conclude that the translator applied indirect translation to the text, instead of direct translation. Gutt (2000 177) defines direct translation as �A receptor language utterance is a direct translation of a source language utterance if and only if it purports to interpretively resemble the original completely in the context envisaged for the original�. Direct translation is offered to the target readers only when the TT readers are familiar with the ST audience�s cognitive environment. This concept of translation assumes interpretive resemblance between the TT and the ST and, as such, refuses to allow any change in explication. In indirect translation, on the other hand, the translator translates the text so that the interpretation of the translation adequately resembles the ST and, at the same time, is relevant to the target audience. In terms of the translation of the short story analysed in this article, because of the linguistic and cultural differences between the TT and the ST, including the TT readers, the examples provided above demonstrate that it was not possible for the translator to achieve complete interpretive resemblance. As such, indirect translation was applied to the text. Therefore, the translator followed a strong relevant approach, making the translation relatively clear to the target audience and, as such, reducing their processing efforts.
Non-relevance related theories of translation encounter difficulties in interpreting translation decisions, whose interpretation depends primarily on relevance-based relationships (Gutt, 2000). In contrast, relevance theory, as demonstrated in the discussion of the above examples, can help to account for most of the translation decisions taken by the translator, in particular when covert translation, which is read as if it is an original text in the cultural context associated with the translation language, is applied (House, 1981). This may lead us an evaluation of the extent to which this theory can help us to understand the translator�s decisions when translating and applying overt translation, a translation in which it is clear that the readers� expectations are not addressed directly, taking into consideration that the target readers� cognitive environment plays an important role in relevance theory (ibid).
