Translation Criticism: A Three-Dimensional Model Based on CDA PDF

# Translation Criticism: A Three-Dimensional Model Based on CDA ## Farzaneh Farahzad ### Abstract This article introduces a three-dimensional model for translation criticism inspired by critical discourse analysis. It assumes that unlike translation quality assessment, which rests on value judgme...

# Translation Criticism: A Three-Dimensional Model Based on CDA ## Farzaneh Farahzad ### Abstract This article introduces a three-dimensional model for translation criticism inspired by critical discourse analysis. It assumes that unlike translation quality assessment, which rests on value judgments, translation criticism needs to explore the ideological implications of translational choices and how they represent identities in target societies. Based on the concept of intertextuality, the model examines the metatext in relation to its corresponding prototext at three levels: the textual, the paratextual, and the semiotic. **Keywords:** Translation criticism, critical discourse analysis, representation, intertextuality, ideology. ## Introduction The terms translation quality assessment (TQA) and translation criticism have mostly been used interchangeably in translation studies, particularly at the early stages of the development of the discipline. The complication might be due to the fact that traditionally, commentaries on translations address not only quality, but also artistic and creative aspects of translations. Moreover, TQA and translations criticism share the same starting point: both assume and are based on some sort of relation between a source text and a target text. However, both terms lack adequate definition in translation studies. ## Redefining the Terms TQA, also known as translation evaluation and translation assessment, emerged as an area in TS to ensure the quality of translations in terms of accuracy, adequacy, naturalness, and other such criteria. TQA starts with a prototext-metatext comparison and ends there. It is restricted to textual boundaries. It makes value judgments and looks for right and wrong /good and bad, within the limits of an ever-changing concept of equivalence. The cultural turn in translation studies marked a shift of paradigm in the 1980s, which in turn called for a shift of focus from what translation is to what translations do. New insights, such as: - the rejection of the concept of equivalence in favor of the concept of 'norms' - the recognition of target texts as target language realities independent of their corresponding source texts - challenging the concept of 'original' in post-structural approaches to translation - representation, identity issues, and power relations in translation - the introduction of concepts such as 'equity' to substitute translational equivalence questioned the nature of the relation between the so-called 'source' and 'target' texts but also opened up a new unexplored area which falls beyond TQA. Translation criticism can look into how the so-called 'source' texts are presented in target societies, what behavior they adopt, what effects they produce in target literatures and societies, how they represent the cultures and identities of the Other, and what ideological implications they bear. ## The Model Translation criticism in this new sense requires: - rethinking of the relation between the prototext (source text) and the metatext (target text) in order to account for the behavior of translations in target societies. - examining translational choices. - identifying the ideological implications of translational choices in relation to their corresponding segments in the prototext. The three-dimensional model presented here uses the concept of 'intertextuality' to redefine and explain the relation between the prototext and the metatext (traditionally called the 'source text' and 'target text' respectively), examines translational choices made by the translator, and analyzes them in the light of critical discourse analysis (CDA) to identify ideological implications. ### Dimension I: Intertextuality The nature of the relation between a prototext and all its existing and possible metatexts, and between the metatexts themselves, cannot be contained in the debatable term equivalence, which, based on the concept of meaning as stable and fixed, considers translation as reproduction and the source text as the origin. One way to explain this relation is to think of it not in terms of equivalence but of intertextuality, which relates texts together, without looking for the origin and sameness. The idea of texts being related to one another was introduced in the 1930s in literary studies, in exploration of literary genres. Bakhtin and Volosinov refer to it as 'dialogue,' 'contextuality,' 'intersubjectivity,' and 'contact between texts.' However, although Bakhtin is known as the originator of the idea, the coinage of the term _intertextuality_ is assigned to Kristeva. To Kristeva, a text is not an isolated piece but 'a permutation of texts' in which 'several utterances taken from other texts intersect and neutralize one another.' She (ibid) believes 'any text is constructed as a mosaic of quotations; any text is the absorption and transformation of another.' As such, a text becomes an _intertext_, referring endlessly to other texts which precede it, and forming part of the chain of other texts to come. It is an heir to the past, a product of the present, and a link to the future. It relates diachronically to prior texts and synchronically to any text being produced in the present. Intertextuality questions the concepts of origin and source. From an intertextual perspective, no text is 'the original'; no text is 'the source' of another. There is no beginning or end to any text, but endless connections and references to other texts. Fairclough sees texts in an intertextual chain as 'part repetition' and 'part creation,' a definition which allows for differences between texts. #### Types of Intertextuality Two types of intertextuality can be distinguished: one is **overt**, in which the intertext bears direct quotations and citations from other intertexts. The other is **covert**, in which the **intertext** relates to and depends on other intertexts not only in terms of genre and discourse but also, and basically, in terms of concepts. The overt-covert classification is by no means dichotomous, for an intertext can relate to other intertexts both directly and indirectly at the same time and within the same textual space. ### Intertextuality in Translation Every specific instance of translation deals with two different languages and linguistic systems, and operates in two different socio-historical contexts. The first thing that happens to an intertext which is translated into another language, is that it gets de-contextualized and detached from its own context and is placed in a new socio-historical context. This movement from one context to another assigns new intertextual properties to it by relating it to the discursive practices of the society it enters. Translation deals with two physically recorded intertexts, traditionally called the 'source text' and the 'target text.' Since within the framework of intertextuality, no text is the source or the origin of any other, the two intertexts are here called the **prototext** and the **metatext**. The prototext is the intertext which gets translated, and the metatext is the result of the act of translation. The two intertexts relate to one another as soon as they are placed in a translational context. The nature of their relationship, as discussed earlier, is not one of equivalence, but may be explained through intertextuality because of the following considerations. #### First, the Prototext: 1. Overly and covertly repeats and transforms other texts preceding it in its own language, in terms of content and form. 2. Is not the source of anything, including the metatext; it does trigger its production, but is not its origin. 3. Does not have an original or fixed meaning; its meaning shapes as a result of what Lodge calls 'the dialogic process between speaking subjects, between texts and readers, between texts themselves.' This process contributes to its multivocity and makes the meaning of the prototext potentially plural. #### Second, the Metatext: 1. Covertly repeats and transforms the prototext in terms of content and form. 2. Is not a reproduction of any other text, including the prototext. 3. Reflects only one of the possible meanings (interpretations) of the prototext. 4. Can never be equivalent to the prototext because it unfolds in a different socio-historical and intertextual context. ### Levels of Intertextuality in Translation In translation, intertextuality operates at two levels: * the local (intralingual) level * the global (interlingual) level #### a) The Local (Intralingual) Level At this level, the prototext relates to all other texts appearing in its own language, the language of the prototext (which is here called the _protolanguage_ in order to avoid the implications of the term source language). It is part repetition in that it repeats the form and content of other intertexts belonging to the same genre and discourse type in the protolanguage and is part creation in that it is distinct from any other intertext as an individual, and new formulation of concepts. Although it induces the act of translating, it is by no means 'the source' or 'the original' since it is in itself an intertext rooted in history, and made up of overt and covert intertextual references to other texts in the protolanguage. #### b) The Global (Interlingual) Level At this level, the prototext is translated and relates through the metatext to all the texts written in all different languages, in terms of content and genre. It appears before its translation(s). So it precedes the metatext(s) in time, while parts of it, (e.g. its content, terms, and formal properties) get repeated in its translation(s). This creates an intertextual relationship between the prototext and all its potentially unlimited number of metatexts which can appear in a given language, the language of the metatext, here called the metalanguage, in order to avoid the implications of the term _target language_ and its inevitable association with the term _source language_. However, every time a prototext is translated, it moves from one intertextual and socio-historical context to another, loosing parts of its properties in favor of those of the new context it is being situated in. ### Dimension II: Critical Discourse Analysis "Critical" means not taking things for granted, opening up complexity, challenging reductionism, dogmatism, and dichotomies, being self-reflective, and through these processes, making opaque structures of power relations and ideologies manifest. "Critical", thus, does not imply the common sense meaning of "being negative" - rather "skeptical". Proposing alternatives is also part of being "critical". One of the most important developments in CDA is a new focus on identity politics ('transition and social change'), language policies, and on integrating macro social theories with linguistic analysis. All these are the present concerns of translation studies as well. In fact, this is what links CDA to translation studies, which since the so-called cultural turn in the 1990s engaged itself with discussions of power in translation. Since then, translating is seen as an act of power and translation as '... a site where discourses meet and compete' (Tymockzco and Gentzler). The introduction of the concept of power into the discipline also gave rise to discussions of representation, identity issues, and ideology, which are the concerns of CDA as well. CDA can therefore serve translation studies by providing ways of looking critically into translations to identify power relations, identity politics, and ideologies. However, following on from Calzada Perez, here the definition of ideology is "not limited to political spheres", but is used in a wider sense to refer to "modes of _thinking_, forms of _evaluating_, and codes of behavior which govern a community by virtue of being regarded as the norm." From among the different approaches in CDA, Fairclough's seems more appropriate for translational purposes. Inspired by Functional Systemic Grammar, Fairclough looks for power relations and ideological implications in linguistic (lexical and syntactic) elements of a text and introduces a methodological analysis of it. This can be used for and in translation criticism. Fairclough suggests an analysis of texts in order to detect ideologically significant lexical and grammatical elements. His analysis can be adapted and used not only for the prototext but also for the metatext, which may be examined both in relation to a corresponding prototext and independent of it. Translationally speaking, two different analysis are possible, one of the prototext and the other of the metatext. Here the prototext and metatext are seen as indicators of ideological positioning of the source and target societies, not of the individual text producers because ideology is collective, not individual. Moreover, the individuals who produce the prototext and the metatext, i.e. the writer and the translator, are seen as agents who unveil these collective ideologies. However, what makes the critical analysis possible is how the reviewer or critic looks into the texts and what s/he looks for. In fact, being critical is not a quality of the text, but of the reviewer and critic. ### Dimension III: Translational Choices Translational choices cover a rather vast area, ranging from lexical and grammatical choices a translator makes, to her/his choice of translation strategies, both at the textual, paratextual, and semiotic levels. #### Methodology In order to implement the model, it is necessary to distinguish between types and levels of translation criticism. #### Types of Translation Criticism There seem to be at least two types of translation criticism: 1. non-comparative 2. comparative #### a) Non-comparative This examines a _metatext_ to detect its behavior and possible effects in the _target literature(s) and society(s)_. The metatext is analyzed and judged as independent of any corresponding prototext, as in the case of works of literature, where the translation gets incorporated in the target language literary polysystem and is treated as what Toury calls a target language reality. This type of criticism draws upon socio-historical factors and basically falls under comparative literature. #### b) Comparative This examines the metatext in comparison to its corresponding prototext. Here the prototext is normally not analyzed for its own sake, but as a means of throwing light on the properties or aspects of the metatext. Such a comparison is not to see if the metatext qualifies as an equivalent to the prototext in terms of right/wrong, but to see whether or not it bears similar/different ideological implications. It looks for discursive strategies and patterns of usage, and examines textual, paratextual, and semiotic features at the same time. The methodology in the present model concerns comparative translation criticism of printed material. However, it can be adapted to non-printed translational material as well, as in the case of dubbed films, localized web pages, and the like. #### Levels of Translation Criticism Comparative translation criticism is carried out at three levels: #### a) Textual Concerns everything in the text, such as: - words - grammatical structures - overt and covert meanings - implications, etc. #### b) Paratextual Concerns everything about the text, such as: - translator's/editor's/publisher's notes - prefaces - judgments - comments - translator's footnotes and endnotes. At this level, everything is analyzed in the light of the socio-historical conditions of production and reception of the two texts. #### c) Semiotic Level Concerns other layers of the text, such as: - the graphic design of the book cover - illustrations - layout - fonts - colors, and the like. At the textual level, lexical choices are checked for their meaning and ideological values. Grammatical choices, (such as agency, modality, tense, etc.) are as well checked for ideological implications. Key concepts in this analysis are that only those translational choices which convey key concepts and/or dominant discursive strategies which are repeated in the inetatext in the form of a pattern are apt to bear ideological implications. #### 1. Textual level Analysis At this level, the metatext is compared to the prototext and analyzed in terms of lexical and grammatical choices and choice of translation strategy(s) and examined for recurrent discursive patterns. ##### Lexical Choices A translator's choice of words is not naïve, because it is not random and follows patterns which affect representations. For instance, a translator who substitutes a cultural item in the prototext by a different cultural item of the _metaculture_ may be doing so to make the _metatext_ more readable, but such a strategy definitely affects representation as well. Lexical choices both denote and imply things, even when they seem to be ideologically neutral. Sometimes a lexical item in the prototext bears an ideological implication, in which case its translation may either have the same or a different implication or may become flat in the process of translation and lose its ideological significance in the metatext. Also, a non-ideological lexical item of the _prototext_ may be translated into an _ideologically significant item_ in the _metatext_, in which case the power relations established in the _metatext_ become quite different from those established in the _prototext_. ##### Grammatical Choices The grammatical forms used in a text assign prominence to agents, actions/events and entities. Fairclough defines this as 'the ways in which the grammatical forms code happenings or relationships in the world, the people or animals or things involved in those happenings or relationships, and their spatial and temporal circumstances, manner of occurrence, and so on.' In text analysis for translation criticism, the prototext and metatext can be compared to detect differences in grammatical forms and identify any possible ideologically significant implications. The categories listed in this section are taken from Fairclough and adapted for translational analyses. However, the list is open to further development. ###### a) Shift of Agency Shifts form a major category here and can be classified both as grammatical choices and translation strategies. Shifts can be obligatory or optional. Obligatory shifts are due to lack of correspondence between the linguistic systems of the _protolanguage_ and the _metalanguage_. Optional shifts are due to translator's choice and may have various reasons behind them, such as stylistic, cultural, or ideological. However, when some optional shifts which concern certain _process-participant formulations_ occur repeatedly as to form a _pattern_ in the _metatext_, they become _ideologically significant_, such as, for instance, temporal shifts (shift of tense/aspect) which affect the temporal sequence of events and the state of affairs. ###### b) Passivization/Activization The passive voice is normally used when 'the action' is prominent, not 'the agent'. According to Fairclough, passive sentences leave "causality and agency" unclear. So, when as a recurrent translation strategy/shift, active sentences in a _prototext_ are translated into passive sentences, agency and causality lose prominence and objects of actions become foregrounded. ###### c) Nominalized Forms In the progress of translation, an event or action expressed by a verb in the _prototext_ may be expressed in the _metatext_ in a _nominalized form_, which is a reduced form, has no tense, has no agent, and is therefore less forceful than a verb. If _nominalization_ is repeated in the _metatext_ so as to form a pattern, then actions become trivialized and are treated as ordinary phenomenon. ###### d) Positive/Negative according to Fairclough, "negation is the basic way we have of distinguishing what is not the case in reality from what is the case." When a positive sentence is translated into a negative, or vice versa, a different aspect of reality is highlighted. This can have ideological significance if it either is used for an item which is ideologically loaded or gets repeated as a pattern in a metatext. ###### e) Tense The change of tense in a metatext leads to a temporal difference and changes the state of affairs. For instance, the use of the present perfect tense in a metatext instead of a past tense form used in the prototext alters the state of affairs by expressing a past event as something still current. ###### f) Coordination/Subordination A common grammatical shift in translation concerns coordination/subordination. Fairclough believes that "in coordination, the simple sentences have equal weight," and that in subordination "the main clause is 'more informationally prominent' than subordinate clauses, with the content of the subordinate clauses backgrounded." When subordination substitutes coordination in a metatext, the information in the main clause gains relative importance over the information in the subordinate clause. This can bear ideological implications if used as a pattern, and moreover affects style. #### C. Choice of Translation Strategies Translation strategies range from 'shifts' to translation methods, (such as literal translation, substitution, omission, addition, foreignization, and the like). All translation strategies have ideological implications and can be viewed from a CDA perspective. One example is omission, which becomes an ideological act of censor when forming a pattern in the _metatext_ and/or concerning certain words and concepts. Another example is the use of 'addition.' As a dominant translation strategy in character descriptions in a novel, for instance, addition becomes significant because it over-represents a character or attributes qualities to it which are not present in the prototext. Addition may as well affect style of expression or even genre. Another example is naturalness. If used as a domesticating strategy, naturalness gains ideological significance because it either overshadows the values and beliefs of the _protoculture_ when translating from a marginalized language into a language of power, or mediates foreign values, those of the _Other_, in the guise of the self when translating from a language of power into a non-power language. ### II. Paratextual Level Concerns whatever is about the text. Any additional information provided along with the _metatext_ falls into this category, such as: - footnotes and endnotes - prefaces, epilogues, and comments by the translator(s) - editor's/publisher's notes and comments Paratextual information, when examined from a CDA perspective, reveal _ideological positioning_. For instance, footnotes which concern the social, historical, and cultural contexts of the _prototext_ are usually given to fill informational gaps which affect the reception or appreciation of the _metatext_ in target societies. Also, the preface or afterword a translator writes to say what translation method s/he has adopted and why s/he has done so can be interpreted critically to detect _ideological implications_. It shows what the translator thinks about the _prototext_, and what position s/he takes for or against it. ### III. Semiotic Level Concerns the visual signs which surround the text. These signs both provide information about the text and function as a mode of representation. They are therefore ideologically significant. They include the graphic design of the book cover, illustrations, and visual images (in web pages, logos in advertisements, etc.) and the like. An interesting example is that of the design of a book of popular psychology which depicts Venus as a symbol of physical beauty on its cover. A bestselling Persian translation of it depicts a sunflower, which signifies nature in the receiving culture and is void of the prototext symbolization. Moreover, the book is introduced on its cover as a 'guide for the newly single', whereas the translation introduces it as 'a guide for married couples'. This becomes an _ideological issue_, not because of whatever reason behind such a tremendous difference, but because the translation addresses a totally different readership in the receiving society and encourages married couples to behave and act like divorced or separated single people of the protoculture. ## Summary This article presents a three-dimensional model that assumes an _intertextual_ relation between the _prototext_ and all its existing and possible metatexts, and between the metatexts themselves. It is based on the assumption that translational choices bear _ideological implications_, particularly when they form patterns in the _metatext_. This model uses insights from CDA, looks into translational choices which make up a metatext, and examines them critically to see what ideological positions they reveal, and how they represent identities. This model comprises a three-level analysis. The first is textual and concerns lexical and grammatical choices of the translator and her/his choice of _translation strategies_. The second is paratextual, and concerns whatever is about the text, such as footnotes, forewords, afterwords, and the like. The third is semiotic and concerns anything surrounding the text, such as the visual signs and images on the book cover, illustrations, etc. The model is designed for written material but has the potential of being adapted for multimodal material (such as dubbed movies, localized games, and theatrical adaptations). ## Works Cited * Allen, Graham. *Intertextuality*. * Bakker, Matthijs, Cees Koster & Kitty Van Leuven-Zwart. "'Shifts of Translation,' in Mona Baker (ed.), *Encyclopedia of Translation Studies*." * Calzada Perez, Maria (ed.). *Apropos of Ideology: Translation Studies On Ideology - Ideology In Translation Studies*. * De Nooy, Juliana. *Derrida, Kristeva and the Dividing Line, An Articulation of Two Theories of Difference*. * Even-Zohar, Itamar. "Polysystem Studies." *Poetics Today* 11.1. * Fairclough, Norman. *Critical Discourse Analysis*. * Farahzad, F. "'Translation As An Intertextual Practice" in Perspectives: Studies In Translatology'*. * Lodge, David. *After Bakhtin: Essays on Fiction and Criticism*. * Simon, Sherry. *Gender in Translation: Cultural Identity and the Politics of Transmission*. * Spivak, Gayatri. "'A Rational for Descriptive Translation Studies,' In Theo Hermans (ed.), The Manipulation of Literature: Studies in Literary Translation"*. * Toury, Gideon. "'The Nature and Role of Norms in Translation,' In: Lawrence Venuti (ed.), The Translation Studies Reader"*. * Tymockzco, Maria and Gentzler, Edwin. *Translation and Power*. * Venuti, Lawrence. *Rethinking Translation: Discourse, Subjectivity, Ideology*. * Wodak, Ruth. "What is Critical Discourse Analysis? Ruth Wodak in Conversation with Gavin Kendall," *Forum: Qualitative Social Research* 8.2.