4.5 Transcription and Translation
Transcribing oral interactions in written form represents a distinctive stage in the process of data analysis (Hutchby & Wooffitt, 2001).It is widely accepted that verbatim transcription can be an ideal database for analysis (Merriam, 2009).Although the process may be tedious and time-consuming, it can be the best way to enable a researcher to familiarise herself or himself with the data.I transcribed the video-recorded seminar interactions and audio-recorded interviews word-byword by following the transcription keys from Eggins and Slade (2004) which offer an explicit and detailed guidance for reading the transcript of classroom discourse.In Table 4.4, the transcription symbols were listed on the left side column and the corresponding meanings on the right side.The phonological and prosodic meanings were not included in the transcription of classroom discourse and interviews because they were beyond the current research foci, yet they effectively guided the process of transcription for better understanding of the verbal meaning.The oversized whole transcriptions of the classroom discourse and interviews were not attached in the book; rather, the representative extracts were embedded in the chapters.
Table 4.4 Transcription keys
(modified from Eggins & Slade, 2004, pp.4-5)
Translation is a vital pre-analysis process to unpack the original language and then to capture and present the ‘conversation literature’ (Eggins & Slade, 2004, p.2).At the same time, it involves a problematic nature of accuracy, readability and politics of representation (Mehan, 1993; Roberts, 1997).A translator will unavoidably bring in his or her own language ideology and interpretation in the process.Nevertheless, ‘there is no such thing as a “natural” mechanism for the representation of speech’ (Atkinson, 1992, p.23).Since the human language is used for communication, arguably it is not necessary to posit the boundary of incommensurability between the languages.In this study, the personal recounts originally in the writing mode of Chinese were translated with the methods of literal (i.e., word-by-word) translation and back translation.The interviews in Chinese, due to the descriptive purpose, were translated clause-byclause.Afterwards, the translation was double checked by a Chinese linguistics researcher in order to ensure the reliability of translation.
In practice of translation, equivalence of meaning would be the most critical consideration.Catford (1965) illustrates how equivalence can be realised at all strata of language, rather than merely at the semantic level, in the book A Linguistic Theory of Translation.Halliday (2001) holds a similar point of view, and adds that ‘equivalence at different strata carries differential values’.Halliday explains how translation can be operated at ranks in an article.In line with Catford’ and Halliday’s points of view, I shall assume that the translated written texts embody sufficient equivalence with the source written texts at lexical, semantic and discourse levels for hermeneutic and data presentation purposes.
As the word-by-word translation for personal recounts was targeted in lexical and semantic-discourse dimensions, I followed Halliday and Matthiessen’s (1999) comparison of some general ideational and semantic features between Chinese and English, which provided a useful guideline for the process.It should be pointed out that there is no necessity to explain the overall grammatical differences between the two languages here, owing to the scope of the present research.Halliday and Matthiessen (1999) explain that language universally has two strata of lexicogrammar and semantics, with a structured lexicogrammatical continuum that is composed of grammatical and lexical features.From this perspective, in most cases, the languages of Chinese and English ‘locate the various semantic domains at roughly equivalent points along the lexciogrammatical continuum’ (Halliday & Matthiessen, 1999, p.299).
However, there are several salient differences between the languages.According to the differentiated grammatical features emerging in my wordby-word translation of personal recounts in particular, I summarised five general disparities between the two languages in accordance with Halliday and Matthiessen’s comparison (1999) as follows:
1） Tense form （linear time）.Chinese is less likely concerned with tense than English， with reference to the linear time in particular； instead， it is constructed through definite or indefinite time adverbs such as ‘once’ （曾经） ‘already’ （好； 了）， ‘last semester’ （上学期）.
2） Deixis.Chinese tends to avoid using specific terms but more general ones.The general word ‘library’ in the Chinese clause 去了几次图书馆 （go + already + few times + library）， for instance， is not used to specify the grammatical meaning of the deixis ‘the’ as in English.
3) Verbal group ^ ‘post-verbs’ (postpositive verbs).Chinese deals with the ‘phase’ of processes more grammatically than English.In Chinese verbal groups, the directional phases, for example, are largely realised by ‘post-verbs’ such as ‘out’ in the example:
4) Nominal groups.When nominal groups need to be identified or quantified, they normally add a measure-type, either collective, partitive, quantitative or individuated before them, such as these underlined words (characters):
5) ‘pre-verbs’ in Chinese prepositional phrases.In both Chinese and English, generally, circumstantial clauses are constructedsimilarly, whereas Chinese prepositional phrases relate more explicitly to the functions of processes.Unlike English, Chinese constructs prepositional phrases by attaching ‘pre-verbs’, analogous to prepositions in English, to figures as a whole, e.g.:
Back translation is a useful method to check the accuracy of the translation and to detect errors in the target translated version (Brislin, 1970, 1980; Douglas & Craig, 2006; Ozolins, 2009), by translating the targeted language (e.g.translated English version) back to the source language (e.g.Chinese).Despite it being widely used in translation-related business, little literature has been reviewed in the fields such as linguistics or academic translation study (Ozolins, 2008).Recent work on the practice of back translation is particularly prevalent in cross-cultural research (Brislin, 1970, 1980; Champman & Carter, 1979; Chen & Boore, 2009; McGor, 2000; Sperber, 2004; Werner & Campbell, 1970), international marketing (Douglas & Craig, 2006) and the medical field (Maneesriwongul & Dixon, 2004; Ozolins, 2009).
However, back translation in this study concerns with linguistic/literal equivalence.To ensure the validation of back translation, ideal translator(s) shall be bilingual and familiar with the literature under study.To this end, I invited a bilingual Applied Linguistics researcher to carry out the back translation, who is Singaporean with English as the first language and Chinese as everyday language.For the operation of the process, I followed what Hewson and Martin (1991) propose as the most adequate one.I initially translated word-by-word the source texts of personal recounts from Chinese into English.For the purpose of presenting data in later chapters, I also cohesively translated the whole texts clause-by-clause.The back translator then independently checked and translated the two English versions translated by me back to the original Chinese language, comparing any inconsistent or mistranslated words, phrases, meanings and cultural gaps.I finally compared both of our translated versions and consulted with the back translator whenever I found any differences in between.