| Author: | Xu, Cui |
| Title: | Identification of L2 interpretese : a corpus-based, intermodal, and multidimensional analysis |
| Advisors: | Li, Dechao (CBS) |
| Degree: | Ph.D. |
| Year: | 2021 |
| Subject: | Translating and interpreting Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Chinese and Bilingual Studies |
| Pages: | xv, 244 pages : color illustrations |
| Language: | English |
| Abstract: | Over the last two decades, there has been an upsurge of interest in the nature of translated language as a form of mediated communication (Baker, 1993). Abundant evidence has shown that translated language manifests some 'universal' lexical patterns that set it apart from non-translated or unmediated target originals, characterized by overall more simplified and conservative language use, an increased level of explicitness, and greater homogeneity among translated texts. As a special form of Translation (Pöchhacker, 2004), interpreting, by contrast, has received much less research attention in this regard due to, particularly, a much more daunting task of corpus construction and compilation. The current research follows the tradition initiated by Shlesinger (2008), and Shlesinger and Ordan (2012), by focusing interpreting per se as both spoken and mediated (i.e., having been translated) language with an aim to isolate interpreting-specific linguistic patterns. Special attention is diverted to simultaneous interpreting into a B language (abbreviated as SI), or L2 interpreting, and the linguistic patterns identified in comparison with non-mediated spoken language (abbreviated as NS) and L2 translation (abbreviated as WT) is thus defined as L2 interpretese. Three major issues are explored, including: 1) What are the general variation patterns of 79 linguistic features under discussion in SI compared with NS and WT?; 2) Are the widely discussed translation-specific patterns also traceable in SI compared with NS and WT based on the current corpus data?; and 3) What are the general co-occurrence patterns of linguistic features under discussion in SI compared with NS and WT? A parallel, intermodal, and (quasi-) comparable interpreting corpus named the LegCo+ corpus has been constructed, featured by one Cantonese component, namely, source speech (abbreviated as ST), and three English components (i.e., NS, SI, and WT) consisting of three pairs of corresponding subgenres, including two types of Q&A sessions, and Debates session. Two-phase data analyses have been carried out based on a selection of 79 linguistic features: an initial unidimensional analysis targeted at the first two broad questions while a multidimensional analysis utilizing exploratory factor analysis (i.e., EFA) addressing the last one. The 79 linguistic features are based on two major sources: Biber's (1988) 67 linguistic features on register variation between spoken and written language; and the much-discussed linguistic features in previous studies on the nature of translated and interpreted language (e.g., Laviosa, 1998c; Sandrelli & Bendazzoli, 2005; Shlesinger & Ordan, 2012). The unidimensional analysis is divided into two main sections: the first section deals with general variation patterns of the 79 linguistic features in SI, NS, and WT, with an aim to isolate SI-specific variation patterns; while the second section zooms in on two widely debated 'universal' linguistic patterns, that is, lexical simplification and explicitation (or increased explicitness as preferred in the current research), both across and within the three English varieties. Results of the unidimensional analysis indicate that: 1) Overall, there are great variations across SI, NS, and WT in terms of the distribution patterns of the 79 linguistic features. SI, in general, is characterized by either an overuse (i.e., more frequent use) or underuse (i.e., less frequent use) of linguistic features associated with more simplified, explicit, and potentially more conventional/conservative language production, compared with NS and WT, while there are other lexical patterns that cannot be readily interpreted, due to lack of information concerning the co-occurrence patterns of the identified linguistic features; 2) In terms of the 'universal' patterns of lexical simplification and increased explicitness, SI in general conforms to the overall patterns of lexical simplification and an increased level of explicitness in relation to unmediated, native spoken language (NS), expect for lexical density (an indicator for simplification) and 'that' adjective complements (an indicator for increased explicitness), which show the opposite trends. These patterns, however, are not always consistent when subgenre comparisons are carried out, indicating a possible genre influence (i.e., the influence of genre types) over the general variation patterns of interpreted language. As far as intermodal comparison is concerned, while SI shows consistent patterns of being more simplified than WT, characterized by lower STTR, higher top 10 vocabulary coverage, lower lexical density, and shorter average sentence length, the linguistic patterns regarding increased explicitness are less clear-cut, and are not consistent across subgenres. The multidimensional analysis also consists of two main parts: general co-occurrence patterns of linguistic features in SI, NS, and WT, with SI-specific patterns being highlighted along different dimensions; and the consistency of SI-specific co-occurrence patterns in terms of subgenre comparisons. The multidimensional analysis reports the following results: 1) Eight factors are extracted, accounting for about 40% of the total variance among SI, NS, and WT, but seven factor are kept in the end; 2) The seven factors are interpreted in functional terms as dimensions, based on the assumption that linguistic features co-occur to realize a shared communicative function (Biber, 1988); 3) SI exhibits specific linguistic co-occurrence patterns along all seven dimensions compared with NS and/or WT, albeit to varying extents. In many cases, SI shares more similarities than absolute differences with NS and/or WT. Dimension 1, 'Involved versus Informational Production', captures potential L2 interpretese as defined in the current research, which shows that SI is more marked in the use of linguistic features associated with involved and informal language production, while unmarked in the use of linguistic features indexing informational and integrated production. This pattern, however, is very genresensitive, as only one subgenre comparison (SI_A vs. NS_A) conforms to this pattern. Dimension 2, 'On-line Information Elaboration with Stancetaking Concerns', reveals the largest variance between SI and NS, but SI and WT show consistently negligible differences and the patterns are very homogeneous across subgenre comparisons, revealing possible shared co-occurrence patterns between translation and interpreting as forms of mediated language due to their more constrained nature in terms of on-line information elaboration, and also possible risk management behaviors of both interpreters and translators (Pym, 2008a). Dimension 3, 'Precise versus Simplified Description', reports distinctive intermodal differences, while the differences between SI and NS are much less noticeable. Along the remaining dimensions, SI shows more similarities with either NS or WT, but given the relatively intermediate dimension scores, the specific linguistic manifestations of co-occurrence patterns among the three English varieties are not very distinguishable. To sum up, interpreted language does showcase distinctive co-occurrence patterns in relation to non-mediated spoken language and translated language from the same source along seven dimensions. However, sometimes these differences are genre-sensitive, and are not equally distinguishable. This research is the first attempt to carry out a systematic study on linguistic variation across interpreted language (SI), non-mediated spoken language (NS), and translated language (WT), from both unidimensional and multidimensional perspectives. The inclusion of a multidimensional perspective, in particular, enriches the existing knowledge about the "multidimensional and multifaceted nature" (De Sutter & Lefer, 2020) of interpreted language, contributing to our knowledge of interpreting as a spoken form of mediated language. The focus on the L2 aspect of interpreted language makes up for the research lacuna due to the lopsided attention on native interpreting, shedding new light on the multi-constrained nature of (retour) interpreting. The fine-grained analysis on (sub-)genre variation on the general distribution or co-occurrence patterns of linguistic features of interpreted language has rich implications for future relevant studies (such as a multifactorial analysis). The findings on the linguistic patterns from either unidimensional or multidimensional perspectives also have implications for interpreter training and teaching. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/11010