Translation Studies 7 Cohesion in translation Krisztina Kroly

Translation Studies 7. Cohesion in translation Krisztina Károly, Spring, 2006 Source: Klaudy & Károly, 2000

Introduction Aim of any aspiring translator: to reflect faithfully the global meaning communicated by the source text Motivation for this investigation: a remark by Neubert and Shreve (1992): ”the argument for a textual approach to translation rests to a great degree on the notion of global textual meaning. It is the global meaning of translation, recontextualized as an L 2 text, that must be matched to the original global meaning of the source text” (p. 139)

This study will argue that one of the means by which this ”match” may be realized is through concrete markers of cohesion identifiable on the textual surface (lexical repetition) aims to demonstrate that the partly revised version of Hoey’s (1991) repetition model, which, via the systematic analysis of lexical repetition capture the macropropositions of texts, is capable of indicating the quality difference between translations.

The analytical tool Starting point = Hoey’s (1991) repetition model the greater part of cohesion is the product of lexical rather than grammatical relations perceives lexical relations as various forms of lexical repetition, serving to show the relatedness of sentences, which produce elaborate patternings in texts sentences are connected if they share at least three points of reference, i. e. three repetition links can distinguish between ”central and marginal sentences”

Types of repetition 2 main types of repetition: (1) lexical repetition (simple and complex) (2) paraphrase (simple and complex) ”lexical items form links, and sentences sharing three or more links form bonds” bonding is a useful tool because it helps identify adjacent or non-adjacent related sentences in texts, and the nets they combine can reflect the organization of the text

Revised taxonomy of repetitions (Károly, 1998, 2002) exempt from the weaknesses of the original model and has been shown to be able to predict quality differences in English written discourse basic unit of analysis = the ”lexical unit” (= a unit whose meaning cannot be compositionally derived from the meaning of its constituent elements)

Lexical unit: (a) one-word units, such as dog, invite, happy, including compounds such as outnumber, teapot, blackbird, darkroom, and lazy-bones, (b) idioms (e. g. , a bitter pill, hit and miss) including idiomatic and phrasal verbs (e. g. , let down, do sth up), (c) phrasal compounds: words often used together to refer to a unique concept composable from the meaning of the individual words in the expression. Examples of this are National Theater, black box, electrical engineer, bank holiday, high winds, conscientious objector, dark brown, and burning hot.

Categories of repetition See handout

Method Texts submitted to analysis Procedures of analysis

Texts submitted to analysis: 3 types (1) an original English newspaper article (i. e. ST) of approximately 250 words (2) its 10 Hungarian translations made by professional translators and (3) its 10 Hungarian translations written by trainee translators

Procedures of data analysis (1) the texts were segmented into sentences and sentences were numbered, (2) the structural components of the text were identified to be able to investigate whether sentences with special discourse function (e. g. , the one which states the "problem") participate in bonding or not: - situation (introduces background material) - problem (the statement of the undesirable condition of things: claim, justification) - solution (statement of the desirable condition, induction) - evaluation (3) the repetition links between each and every sentence were identified and classified, (4) repetition bonds between sentences were identified, (5) repetition links and bonds within the texts were counted. + inter-coder reliability checked + t-tests

Sample analysis See handout

Results Quantity of various types of repetition The repetition use of professional translators resembles that of the ST more than the novice translators' use (especially simple repetition) Reason: ST is a highly factual, informative type of text, to reflect faithfully its global meaning, the translator cannot avoid the verbatim repetition of particular lexical units, the same way as it is done in the original English text

Results cont. The combination of repetition links and bonds (1) professional translations' mean values are closer to the ST's values, and the novice translators differ more considerably professional translators "weave" the text via the use of repetitions and bonding in a more similar way to the ST than novice translators do (2) t-tests: the two groups of translators differ in their strategies of using intersentential links and bonding the number of links and bonds + the number of central sentences are significantly higher in the professional translations than in the work of novice translators

“Repetition matrices” density of bonds: professional translations (similarly to the ST) novice translations amount bonds among key sentences: professional translations (similarly to the ST) novice translations title: professional translators seem to integrate the title more heavily into the text via bonding (as done in the ST) ***