Power and Hegemony in Research Supervision : A Critical Discourse Analysis

This qualitative case study was aimed to investigate the notions of power and hegemony in research consultations around the construct of language ideologies. The study employed purposive sampling by selecting two MS (Education) research supervisees with their supervisors. It identified a social wrong of power and hegemony in research supervision practices with a transformative spirit, and aimed to bring about awareness and modifications in research supervision. Methodically, the study followed the analytical approach of Norman Fairclough’s Critical Discourse Analysis (CDA, 1989; 2003). Data were collected through observations and transcriptions of research consultation meetings and interviews of the participants. The textual analysis of case one revealed how language became a source of establishing asymmetrical power relations between the participants. Contrastingly, for case two, the consultations were not hinged on language aspects and had more equality and homogeneity between the partners. The findings of the study indicated that language ideologies played a very important role in sustaining the dominant position of the powerful agent, that is, the supervisor. Based on the findings, this research will contribute towards implementation of better democratic practices of supervision, equity and emancipation in supervision as a social practice.


Introduction
The contribution of the field of language ideology is overarching in linguistic anthropology, linguistics, sociolinguistics, pragmatics, and discourse analysis.The field has contributed immensely in understanding the use of language in society, and has developed in the 20 th century as an independent area of study, combining perspectives from linguistic ethnography and the social scientific study of ideology.Language consists of deep sociocultural, political and historical processes, and as such, the very existence of language is ideological, which involves power, authority, and control (Blommaert, 2006).Language ideology has attained a significant place in the ethnographic tradition of linguistic anthropology, variationist sociolinguistics, and applied linguistics.Major scholarships in the field are concerned with the ideologies prevalent in education, gender studies, globalization and political economy, media and organizations.Some of the major probes have been related to power and control, peripheral normativity, racism, inequality and child labour.
Language ideology refers to a shared body of common beliefs, views and perceptions about language which includes cultural assumptions about language, nature and purpose of communication, and patterns of communicative behaviour as an enactment of a collective order (Woolard, 1992).According to Tollefson (1999), language ideology tries to capture the implicit, usually unconscious assumptions about reality that fundamentally determine how human beings interpret events.Inspired mainly by the Marx-inspired works of Bakhtin and Voloshinov in the 20 th century, contemporary interest of language ideology has been in studying aspects such as dialogue, voice, social struggles, and social contests.More importantly, language ideology has contributed significantly to the field of education in the ethnographic tradition of linguistic anthropology (Cazden & Hymes, 1972;Heath, 1983;Mehan, 1979;Philips, 1983).There have been research studies related to language ideology and identity in literacy (Levinson, 2005) and classroom practices (Rogers & Christian, 2007).Research shows that on the basis of language ideologies, hierarchies are built in language classrooms, both within languages (De Costa, 2010;Carreira, 2011) and between languages (Luykx, et al., 2008;Flores & Murillo, 2001;Mori, 2014;Volk & Angelova, 2007), cultural variability in education (Blommaert, 2006), for example student diversity in terms of their privileged or under-privileged educational backgrounds.

Purpose of the Research
The purpose of this qualitative case study was to investigate the notions of power and meaning making between research students and their supervisors' discourses in research consultations in higher education.The study aimed to explore the critical discrete moments in discourses in which power and domination were built discursively by the participants around the theme of language ideologies.The uniqueness of this study is the depth of specific understanding of literacy processes and practices in this specific context.There was no attempt at generalization for a larger audience and for organizational studies.

Rationale of the Study
This study's premise is based on Woolard's (1992) critical stance of ideology which is directly linked to attaining social, political, and economic power.This critical aspect of ideology seeks to critique and improve research and inquiry along with the social aspects.Language ideologies look at the deep layer of inequality in society and explore how one variety or one language is considered as standard language and the others are considered substandard.It investigates why and how qualifiers like 'good' or 'bad' or 'worse' are associated with particular varieties.Additionally, language ideologies have a wider ideological uptake, as failing to meet such norms equals failure in society, education, academics, business or employment domains.Due to the globalization and hegemonic control by the western colonies, English has been used in the domains of power, of media, courts, education, and politics.It is the standard language which attained its status as the language of power.Standard English is only variety of English which had been identified as standard by the users of English who were in power.

Significance of the Study
A study such as this one is significant because of three main reasons.Firstly, the emerging scholarship in Pakistan has recently led to the increased participation of students in higher education.These students are the products of the educational system of a country which records ambivalent language policy planning and practices.The study contributed to the existing knowledge on research consultations in higher education in Pakistan by pronouncing the ways through which inequality in systems of education and medium of instruction led to the forming of certain language ideologies in academia.Secondly, such a study Power and Hegemony in Research Supervision would be able to provide invaluable insights for language policy planners regarding the challenges related to the ideologies of English language that the students face in research consultations in higher education.Thirdly, the study tried to bring to the forefront the power of discourse in stimulating major ideological changes in society; since academic institutions are part of society, organizational cultures are affected by such ideological affects.

Research Question
The study was conducted at a private university of Pakistan to investigate the language ideologies prevalent in research supervision practices of two MS (Education) research supervisees with their supervisors.This paper reports on the linguistic analysis of the structure, forms, and meanings of the text, that is, microanalysis using the texts of research consultations.The central research question of the study was: What aspects of power and hegemony are embedded in students' and their supervisors' discourses in higher education around the construct of language ideologies?

Literature Review
In order to investigate the research question, literature review was conducted related to language ideologies, scope and impact of ideology, critical discourse analysis and power and hegemony in literacy practices.Discourses are heavily affected by language ideologies that thrive among a community or group of people concerned.Language ideologies are beliefs about language which have become a part of social processes.According to Bourdieu and Dubois (2007) (Language ideologies are) "A set of beliefs on languages or a particular language shared by members of a community … These beliefs come to be so well established that … (they) end up being 'naturalized', or perceived as natural or as common sense" (p.104).Language ideologies are conceptions such as status, function, norm, and ownership of a certain language.They have an effect on language planning that is built on societal desirable forms of language use and are reproduced by the members of society using it (Blommaert, 2006;Spolsky, 2009;Woolard, 1992).

The Scope of Language Ideologies
The scope of language ideology is vast.It grew out of linguistic anthropology and inquires into the link between language and culture, affecting the works of linguistic scholarship and the larger socio-political and cultural-historical schemes in which it could fit (Calvet, 1974;Errington, 2001).It is mainly based on Marxist theory which explores the mediation between material and ideational aspects of reality.According to Marx, dominant ideology is the dominant bloc in any society, controlling the means of mental production, such as religion, education, and communication.The acceptance of dominant ideas by the dominated creates what Marx referred to as 'false consciousness', that prevents the dominated from seeing the world as it really is.Abandoning the notion of 'false consciousness ' Althusser (1971) redefined ideology as a set of cultural practices facilitated through the religious institution, the educational system, the family, and the law.The main function of these cultural practices, Althusser explains, is the division of individuals into groups to give them systematically different treatment.

The Impact of Language Ideologies
Different linguists have tried to interpret ideologies according to their own way, but a common thread which runs among all the concepts of language ideology is the impact of language ideologies on every aspect of communication.In its affect, language ideologies are responsible for creating classes and groups among people.It even impacts linguistic forms and functions as well as the wider behavioural settings in which they occur (Blommaert 2006;Woolard 1994).According to Tollefson (1999), language ideologies try to capture the implicit, usually unconscious assumptions of people about the way they interpret events.Whorf's (1897Whorf's ( -1941) ) proposition is that grammatical organization is cultural and social (Blommaert 2006).Taken in this manner, the very names of languages are ideological like Urdu, English, or Italian, as languages have evolved through deep and long historical processes.

The Critical Stance of Ideology
The study is based on the critical stance of ideology (Woolard, 1992).Critical stance of ideology studies the aspects related to the process of acquiring power and authority socially and politically.This critical aspect of ideology seeks to critique and improve research and inquiry along with the social aspects.Major Power and Hegemony in Research Supervision works include those by Balibar (1985Balibar ( , 1991)); Bourdieu (1982); Crowley (1989); Finegan (1980) and Joseph and Taylor (1990).

Critical Discourse Analysis
The discourses of the participants of the study were analysed using the method of Critical Discourse Analysis (CDA).The CDA method proved invaluable as it provided insights into the relationship between language, power, and society (Gee, 2005;Rogers, 2004).The scope of CDA is not only language-based, but its concern lies with unveiling patterned mechanisms of power asymmetries.The study is based on Norman Fairclough's CDA method of analysis.His approach aptly fitted the design of this study as it has the scope and capacity to critically examine higher education students and their supervisors' dialectically formed discourses during research consultation meetings.The approach and method provided insights to the study to investigate various social, cultural and institutional affects along with the linguistic elements that were operationalized in students' and their supervisors' dialogues.According to Fairclough (1989), power in discourse is controlled by powerful participants, who are the ones "controlling and constraining the contributions of non-powerful participants" (p.38-39).Research supervision generally involves a discursive relationship between the supervisor and supervisee, as they are participants who are continually processing meanings and negotiating aspects of power and identity in their texts.In Pakistan, research supervision is quite an autonomous field, as different academic supervisors have a vastly different style of research supervision.CDA's critical stance on literacy brought invaluable insights to this study by analysing the discourses of power and hegemony in research consultation meetings.

Methodology
Two MS (Education) students and their supervisors were selected as participants for this qualitative case study.The methodological framework of the study is based upon the three levels of Norman Fairclough's Critical Discourse Analysis (CDA) (1989; 2003) which include: Text analysis (description), processing analysis (interpretation), and social analysis (explanation).This paper focuses on the textual analysis based on the description of the textual linguistic features of language ideologies manifested in research supervision in which language becomes a source of power and hegemony.
The data were collected using observations and transcriptions of research consultation meetings and individual interviews with the participants.CDA does not follow a fixed, regimented way of gathering data; accordingly, the researcher tried to be flexible in order to include any phenomenon worth attending to.Besides, data analysis started along with the collection of data, which was a circular process, moving from data collection to interpretation in a circular manner to explain the theory.A verbal as well as a written consent was taken from all the participants of the study.The participants were assured of the confidentiality of the responses and the data and all data were transcribed in complete seclusion.
Fairclough's approach to CDA (1989;2003) was adopted, which consists of three inter-related processes of analysis tied to three inter-related dimensions of discourse.These three dimensions are: 1.The object of analysis 2. The processes by means of which the object is produced and received (writing/speaking/designing and reading/listening/viewing) by human subjects, and 3.The socio-historical conditions which govern these processes.
Each of these dimensions required a different kind of analysis, which is: 1.Text analysis (description), 2. Processing analysis (interpretation),

Social analysis (explanation).
This paper focuses on the first level of analysis that is description.At this level of analysis, the focus of CDA is on textual-linguistic features of data, such as grammar, vocabulary, cohesion, etc.By studying the forms of language, one can discover the social processes and also the specific ideology embedded in them.For linguistic analysis, Fairclough draws on Halliday's Systemic Functional Grammar (SFL).SFL looks at linguistic categories of grammar and genre as social functions.Following Fairclough (2003), a relational view of discourses was adopted.There were several layers of analysis and an attempt was made to look at the relations Power and Hegemony in Research Supervision between different levels of the categories, based on the transcripts of discourses (consultation meetings between the research participants and their supervisors).

Figure 1. Discourse as text, interaction and context
As is shown in Figure 1, the text is at the central level of analysis and examines the textual-linguistic aspects to understand power and hegemony.CDA is strongly based in theory, and faces the challenge of mediation between theories as applied to larger society and discrete moments of social interaction which takes place in texts (here transcripts of discourses).The peculiarity of CDA designs is its flexibility of approach and method of analyses.The data were analyzed using multiple theories and approaches, depending upon the unfolded phenomenon.The transcripts were read a number of times to select and code the data.

Figure 2. Discursivity in data
As is shown in Figure 2, the research method observed recursivity in the data collection procedure.

Findings
The study took the critical stance of ideology based on Woolard's (1992) premise, and its purpose was to investigate the notions of power and hegemony around the construct of language ideologies in research supervision practices.The study designed its conceptual framework from Fairclough's Critical Discourse Analysis (CDA) (1989; 2003) approach and method.The aim of CDA is to unveil latent (language) ideologies, contested relations of power and authority, and issues of inequality and oppression in society.According to Fairclough and Wodak (1997), the basic tenets of CDA are the view of discourse as a means of addressing social problems; acknowledging power relations as discursive; and constituting a free and democratic society.For the sake of studying commonalities and differences, case one and case two are given separately under the following headings.

Auxiliary Verbs
Case one.The auxiliary verbs are a way of studying power and authority in the way that obligation, possibility, suggestion, probability or definitiveness of action are expressed in interactions.It explains the asymmetrical power relations between the participants.Taking its impetus from Fairclough's (2003) framework, the subjects and verbs were separated to find the auxiliary verbs which express.The pattern of auxiliary verbs showed more directness in supervisor's statements owing to her powerful position: I want to, I am, I want it this way, It should be used, I have explained, I have seen.The supervisor used exceeding number of definitiveness in action, and from the beginning of the meeting, this set the tone of the interaction.This directness was mainly due to her command of the situation as the supervisor.Her command of English also added to her confidence level.
Case two.The case two participants used more general way of giving instructions or directives and making suggestions.There were very few auxiliary verbs used in using the functions mentioned above.Instead of I and you, there was more use of passive verbs by both the participants.The supervisor preferred giving examples of the mode of operation and used we and us.

Grammatical Mood
Case one.There were a number of declaratives, negatives and interrogatives used by the participants which set the tone of the conversations.The supervisor used a number of declaratives statements.
There were also imperatives with a tag question at the end like 'ok', 'alright', 'yeah?' in the directives given.Bach and Harnish (1982) explain that directives "express the speaker's attitude towards prospective action by the hearer," (p.47) and suggest different types of directive sub-classes: Requestive; requirement; prohibition; permissive and advisory.Also according to them, when speakers make a requirement speech, that is a speech act requiring the interlocutor to take some action, they are presuming that they have authority over the hearer.There were also a number of negatives used in the text, in which the supervisor refused to accept the supervisee's assertions.There was an element of absoluteness in these negations.The orders of discourse directed the speech to be straightforward, without any surge for explanation.The supervisor also used a number of interrogatives.This was also ideational as the supervisor could ask a series of questions, allowed by her powerful role.In the meetings, open-ended questions and yes/no questions along with the tag questions were used.However, the supervisee rarely came up with longer stretches of explanation.
Case two.Affirmatives and declaratives were used in the discourses of the supervisor, with a minimum of negatives and interrogatives.There were tagquestions used, like alright, ok.The text was largely structured through narratives and explanations, comparing strategies of research adopted by other people and using one's critical thinking to decide which approach in research should be taken.The supervisor used a fibula story narrated for a particular aim in his conversation.According to Fairclough (2003), fibula is a series of events presented in different chronological order.The supervisor used a combination of direct and indirect reporting of his academic life from which he took inspiration to motivate the supervisee who sounded in low spirits in the early part of the meeting.

Pronouns
Case one.Personal pronouns were used by the participants.There was an excessive use of 'I' and 'you' pronouns by both the participants.There was also an occasional use of 'we' and 'they' in references to Pakistani teachers and students.
The use of personal pronouns for the supervisee seemed quite direct.The gap in politeness could be explained through Brown and Levinson's (1987) model of politeness where they talk about the relative power of the senior which demands a high level of politeness from subordinates or students to seniors or supervisors.In this context, while the use of personal pronouns may be appropriate from the senior, it may be reckoned as unsuitable from the supervisee keeping the politeness aspect in view.
Case two.The personal pronouns 'I' and 'you', and 'we' were largely used by the supervisor.
"I mean you cannot go exactly as per this, in it we will find that from the discourse unit only...." The tone was homogenous and democratic in giving instructions to the supervisee.The supervisee did not use personal pronouns with the supervisor and talked in more general sense.

Lexical Items
Case one.The lexical items related to technical vocabulary of ICT were used excessively in the discourse: Video blogs, blended learning and teacher driven action research.This was the supervisee's domain and here the inherent power struggle could be seen where the supervisee was in the main position and guided the supervisor.The discoursal element of power was high in these exchanges where the supervisee's indicated her control on the topic.Since the supervisor and supervisee constructed their domains where they were in powerful positions, their struggle for power was contested in very soft ways between them.Supervisor's domain was academic English, and by virtue of being an English language teacher throughout her career, her adherence was more towards language structure and writing skills.
"Then it's identifying, identifying resources, ohh okay/ now I got it alright so it's identifying resources." The supervisee used an incorrect expression 'identify' instead of 'identifying' which created confusion.The ideological base of such an assertion Power and Hegemony in Research Supervision made the supervisor appear more powerful.Her English language skills were a strong aspect of her identity, which left the supervisee in a more disenfranchised position.The supervisee's domain was ICT and she tried to make her mark in her input, still it left her in a less powerful position.English has become one strong indicator of membership in lower, middle or upper class strata of Pakistani society.While a lack in English is interpreted as the low socio-economic levels and poor education standards, high socio-economic group is associated with better standards and elegant style of living.Thus, English language is the most obvious indicator of status, and has become the structured norm in education and social practices.In this sense, English is responsible for creating class differences and may become an exploitation tool for power, authority and control.The political history of the country evidences how English has been controlled by the political agencies, by allowing access of English education to few and denying it to others.As a case in point, the speech repertoire of the supervisor was rich, whereas the supervisee used English in disjointed phrases most of the time.
Case two.The participants' discourses were textured by lexical items pertaining to knowledge, critical thinking, goals of education, research writing, thinking out of the box, and mainly by Islamic references and Muslim scholars.While the lexical items commonly pertained to research and writing, the discourses and the lexical items largely differed in its orientation to Islamic lexis and knowledge based expressions.English was regarded a problem, a stigma in the discourse, as the supervisee expressed: "I don't have the problem, how the mind has to be used and what is to be done ... it is the problem of reading that English...." The notable aspect was the ordeal of 'reading that English'.The other aspects of research writing were not a problem for him as it was only the English element that was challenging.Ideologies related to language have a dominating role to play in work related tasks as people are either efficient or deficient in English.In this way, language ideologies have a vital role to play in people's lives.

Internal Relations
Case one.Internal relations consist of paradigmatic features of language structures, continuity in sentences and use of coherent devices.These aspects were also of absent in the speech of the supervisee.The supervisee used phrases or broken sentences and used only a few coordinating or subordinating sentences.This aspect related to language ideologies, as she might have avoided speech because of the limited language command.It may be that she got overwhelmed by the supervisor's command over the language and owing to this she chose to be silent most of the time.According to Fairclough (1989), silence works as a tool for the lesser powerful as they can refrain from being corrected or checked.Considering the gap, the supervisor adopted a more powerful stance.
Research is becoming a commodity in education, as an artefact that could be possessed, taken control of and made use by the powerful.Language ideologies serve the purpose of understanding language as a social practice.Referential ideology of language has contributed in understanding language in its socio-cultural context (Silverstein 1979(Silverstein & 2006)).According to Blommaert (2005), the cause of linguistic and social inequality in society is because of the incompetence of the speaker to perform at a desirable level in society.Bourdieu (1971) and Bernstein (1971) explain that language exercises power through prestige and stigma in language.
Case two.The consultations were marked by its genre mixing with narration, argumentation, compare and contrast and cause and effect relationship.

"I had gone out of the way to uhhh give credibility to my research so that is what I believe and that is what I think I mean this is what my approach is. I would like my students uhh uhh my supervisees to have a very reliable research / a thorough research".
There were continuous references to three aspects during his interview.The first was a criticism of the regimented structure of thesis writing, and English was equated to structured norms in writing; the other reference was a positive one related to knowledge aspect; and the third reference which was a strong one, was related to Islamic education.It would be discerning to note that while its structured aspect had a negative import, knowledge and Islamic education had positive connotations in the text.the meetings and the interview.

Conclusion and Recommendations
Collectively, the research question was able to yield interesting findings on the textual analysis of the research consultations, and helped in developing better understanding of the notions of power and authority in literacy practices.Following Woolard's (1992) critical stance of ideology, the research question played a vital role in investigating how discourses during the consultation meetings became the source of power relations between the supervisees and their supervisors.The linguistic analysis of the consultation meetings revealed how language assumes a manipulative tool by the powerful agents in literacy processes, which become a source of hegemonic control in academia.The case one discourse of the participants was largely affected by the institutional structure and control, which influenced the supervision practices and gave them a highly structured form.Language was a major concern in the consultation meetings, which became a source for establishing asymmetrical power relations between the participants.
Case two was in a sharp contrast to case one in approach and style of supervision.The structure of supervision was very flexible and there was no constraint on use of the English language.The findings reveal an analytical approach toward the philosophical debates on a Muslim scholar.The ideological index was very high as the supervisee's preferred that the language of communication remains Urdu and he was given this autonomy and freedom by the supervisor to do so.The artifactual view of language was strongly opposed by both the participants, and was taken as a constraint in the way of conducting research and acquiring knowledge.The voice of the supervisee was given strength and encouragement by the supervisor.
On the basis of the findings, the study recommends that English language should not be considered as a single criterion for communication in research consultation meetings.The study demonstrated that if the language barrier is removed, a lot of critical and creative thinking can be inculcated among the supervisees of research.A pluralistic approach should be adopted where not one language but bilingualism or national language should also be encouraged.