Identifying and improving reading comprehension in the translation process : a visualisation approach

Nwachukwu, Jeremiah Felix (2017-12)

Thesis (PhD)--Stellenbosch University, 2017.

Thesis

ENGLISH ABSTRACT: The experiment described in this dissertation is based on three basic assumptions, all of which were derived from research findings providing explanations relevant to the requirements of translation as performance process (Lörscher 2005; Wolf 2017). The first assumption is the obvious fact that “success in translation is predicated upon an ability to operate literately in more than one language” (Malmkjær 2004:4). This at least underscores the continuous relevance of translation in second-language learning and vice versa. The second has to do with the recent general acceptance among translation scholars that the type of reading involved in translation is more challenging than the one meant for other forms of communication (Jakobsen & Jansen 2008; Scott 2012:15; Boase-Beier 2014:214). Finally, considering the special nature of reading for translation, it has become expedient to identify certain skills that can enhance source-text (ST) comprehension in translation and to teach these skills to trainee translators (Hurtado Albir, Alves, Englund Dimitrova & Lacruz 2015:17). The methodology of the present study involved the recruitment of 14 students of French who were divided into two groups of seven each. Group A members initially received training on the use of Sowa’s (1984) conceptual graphs (CG) formalism in the analysis of selected news articles for the purpose of translation. CG is a visualised graph grammar derived from a phrase-structure grammar. The rationale behind the use of this formalism was based on the relationship identified between CG and the cognitive linguistic principle of flexibility and the dynamic nature of language as a meaning-making tool to enhance text visualisation. The experimental process involved the participants reading a ST written in French, answering a number of text-comprehension questions about it and translating the text. The performance of these tasks was monitored with the aid of Flashback®, a screen-recording software program that performs video and audio recordings of both on-screen and off-screen activities in order to examine how the performance of one group differs from that of another. The analysis was based on three research questions and seven hypotheses, which compared the two groups on the quality of ST comprehension in their translations, task time and dictionary use. A visualised presentation of the results reveals outcomes indicative of an overall tendency towards better ST-comprehension performance by Group A than by Group B. Tasktime results show that the group trained in the use of CG completed the experimental tasks relatively quicker than the other group. On dictionary use, a review of the process data shows that the untrained group had more dictionary lookups than the group trained in text visualisation. The conclusion is that, since training might hold a key to the development of expert tendencies in student translators, the use of text visualisation as a tool is advocated.

AFRIKAANSE OPSOMMING: Die eksperiment wat in hierdie proefskrif beskryf word, is gegrond op drie basiese aannames. Al drie is ontleen aan navorsingsbevindings wat verduidelikings verskaf wat relevant is indien vertaling as ’n tipe uitvoering beskou word (Röscher 2005; Wolf 2017). Die eerste aanname is die ooglopende feit dat sukses in vertaling op die vermoë berus om in meer as een taal geletterd te funksioneer (Malmkjær 2004:4). Dit benadruk minstens die voortgesette relevansie van vertaling in tweedetaalleer en vice versa. Die tweede aanname het te make met die onlangse algemene aanvaarding onder vertaalteoretici dat die tipe lees betrokke by vertaling uitdagender is as dié wat vir ander kommunikasievorme bedoel is (Jakobsen & Jansen 2008; Scott 2012:15; Boase-Beier 2014:214). Ten laaste, in ag genome die spesiale aard van lees vir vertaling, is dit raadsaam om spesiale vaardighede te identifiseer wat bronteksbegrip tydens vertaling kan verbeter, en om hierdie vaardighede aan leerlingvertalers oor te dra (Hurtado Albir, Alves, Englund Dimitrova & Lacruz 2015:17). Die metodiek van die huidige studie het die werwing van 14 studente behels wat Frans aanleer. Die studente is in twee groepe van sewe elk verdeel. Die lede van groep A het aanvanklik tydens die ontleding van gekose nuusartikels wat vir vertaling bedoel is, opleiding in die gebruik van Sowa (1984) se formalisme vir konseptuele grafika ontvang. Konseptuele grafika is ’n gevisualiseerde grafikagrammatika wat afgelei is van frasestruktuurgrammatika. Die motivering vir die gebruik van hierdie formalisme is gegrond op die verhouding wat geïdentifiseer is tussen konseptuele grafika en die buigsaamheid en dinamiese aard van taal as kognitiewe linguistiese beginsel om as betekenisskeppende instrument vir die versterking van teksvisualisering as metode gebruik te word. Die eksperimentele proses het behels dat die deelnemers ’n bronteks (BT) lees wat in Frans geskryf is, ’n paar begripsvrae daaroor antwoord en die teks vertaal. Die uitvoer van hierdie taak is gemonitor deur middel van Flashback®, ’n skermopnameprogram wat video- en oudio-opnames maak van aktiwiteite wat op en weg van die skerm plaasvind om sodoende te ondersoek hoe die prestasie van een groep van dié van ’n ander verskil. Die ontleding is gegrond op drie navorsingsvrae en vyf hipoteses wat die twee groepe vergelyk wat betref die gehalte van BT-begrip in hulle vertaling, taaktyd en woordeboekgebruik. ’n Gevisualiseerde aanbieding van die resultate toon ’n uitkoms wat dui op ’n tendens van beter BT-begripprestasie deur groep A as groep B. Taaktydresultate toon dat die groep wat opleiding in die gebruik van konseptuele grafika ontvang het, die eksperimentele take relatief gouer as die ander groep voltooi het. Wat woordeboekgebruik betref, toon ’n oorsig van die data dat die onopgeleide groep woordeboeke meer dikwels geraadpleeg het as die groep wat in teksvisualisering opgelei is. Die gevolgtrekking is dat, aangesien gespesialiseerde opleiding moontlik die sleutel tot die ontwikkeling van leerlingvertalers inhou, die gebruik van teksvisualisering as instrument aanbeveel word.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/102741
This item appears in the following collections: