4.1　Introduction

Genres are regarded as social actions which are performed to achieve certain communicative purposes through speech or written texts(Bhatia,1993;Miller,1984;Swales,1990).Genre analysis has been used to investigate the communicative purposes and schematic structure of different genres,typically for pedagogical purposes.Previous genre analysis studies roughly fall into two categories in terms of move identification.The first type of study classifies moves through text analysis,relying on the analysts’understanding of a genre(e.g.Swales,1981;Nwogu,1997;Yang &Allison,2003;Piqué-Noguera,2012;Önder,2013;Samraj &Gawron,2015;Labrador,Ramón,Alaiz-Moretón &Sanjurjo-González,2014;Samraj,2016;Nathan,2016).Researchers following this approach typically read the texts multiple times to develop an appropriate move model.In some studies,linguistic cues are used to assist in the identification of moves(e.g.Labrador et al.,2014).The second type of study identifies moves by combining text analysis with input from experts in the relevant field(e.g.Lim,2006;Pryor &Woodward-Kron,2014;Yang,2015;Breeze,2016;Lee,2016;Chiu,2016).Studies using this approach either first consult specialist informants to better understand a genre,and then use the informants’view on the communicative functions of a genre to inform text analysis(or perhaps even have informants participate in the text analysis.e.g.Basturkmen,2012),or develop move models through text analysis and then check the model against the understanding of experts(e.g.Lim,2006).

As for studies taking the first approach,analysts read the text closely to identify the different communicative functions.Since researchers are often outsiders to the discourse community under examination,their understanding of rhetorical functions can differ from that of community insiders.It is necessary to distinguish the perspective in genre analysis made by community members(i.e.the emic view)and researchers(i.e.the etic view)since the difference is related to the type and nature of knowledge discovered through scientific investigation,and thus lies at the heart of the epistemological foundation of genre analysis.

The emic knowledge of genre practice is associated with community members’view on the functions and content of a genre used in their community,whereas the etic knowledge is the scientific observers’understanding of the functions and content of a genre.Emics(or emic knowledge)refers to mental or behavioural systems that are regarded as appropriate by participants(i.e.insiders),while etics(or etic knowledge)refers to the systems developed by the community of scientific observers(i.e.outsiders)(Headland,Pike,&Harris,1990).In situations where it is appropriate,scientific research combines both emic and etic knowledge to obtain a comprehensive view on human behaviour and cultural phenomena(Lett,1990).

The importance of integrating the insider’s view in genre analysis has been acknowledged(Askehave &Swales,2001;Bhatia,1993).Askehave and Swales(2001)indicate that the communicative purposes of genres cannot usually be discerned straightforwardly and the process of identification often involves ethnographic work and is therefore labour intensive.Consulting specialist informants to identify or to check the communicative purposes and rhetorical moves in genre analysis is a recommended practice(Bhatia,1993)and has been widely used in the second type of research(Basturkmen,2012;Lim,2006;Paltridge,Starfield,Ravelli,&Tuckwell,2012;Swales &Rogers,1995;Swales,1998).Although Swales contends that“active discourse community members tend to have the greatest genre-specific expertise”(Swales,1990,p.54-55),he also suggests that expert members may hold different views on the purpose of a genre(Askehave &Swales,2001).Indeed,in interviews,respondents are required to provide responses that are necessarily spontaneous and therefore may not be accurate since participants have limited time to make a judgment.It is not surprising that for a single communicative function the respondents’opinions may vary and be incomplete.Analysts therefore need a more reliable and systematic way to gain an insiders’understandings of communicative functions of a genre.

The present study proposes to make use of metatexts,i.e.,the language used by writers to navigate the reader through a text,to obtain the writer’s own perspective on communicative functions.My preliminary investigation found that thesis writers not only signpost the content of a section using metatexts,but also explain the communicative purpose behind the inclusion of particular content,which is consistent with the observation of Mauraunen(1993)and Bunton(1999)who also found that writers not only use metatext to describe the organisation of main texts but also comment on it.

To demonstrate the approach,the researcher analysed the metatexts in 30 Master’s Theses Introductions and Literature Reviews to investigate the rhetorical structure of these sections.Swales(1981,1990)developed the CARS model of Research Article Introductions,which,along with subsequent revisions,remains the foundation for a wide range of genre analyses.The model has been used in the analysis of Introductions and Literature Reviews of Master’s and PhD theses,which is a genre different from RAs,with different target audience,communicative purposes and length.Although some new steps have been added,the existing models of Thesis Introductions and Literature Reviews(Bunton,2002;Kwan,2006)closely resemble the CARS model.The researcher intends to use metatext analysis to re-examine the schematic structure of Thesis opening section to explore writers’view on the communicative functions of the part-genre.This study addresses the following research question:

·What is the thesis writers’view on the schematic structure of the Introduction and Literature Review chapters of Master’s Theses?