By Azeema Anhar
For this issue of MUSE, Azeema Anhar interviewed Dr. Mahmud Hasan Khan, one of the most cherished faculty members at DEH. In this interview, he reflects on his journey that begins in literature and gradually moves into discourse analysis and language policy, eventually finding its ground in the lived linguistic realities of the Chittagong Hill Tracts. Along the way, he draws out how questions of language are never separate from identity, power, and the quiet politics of everyday life.
Interviewer: How did your journey from discourse analysis to sociolinguistics and language policy begin, and what first sparked and later shaped your interest in indigenous languages in the Chittagong Hill Tracts?
Sir: I was a student of English literature at Jahangirnagar University. I completed my BA and MA there. After that, I went to Malaysia for another MA. In Malaysia, there is no proper department of philosophy and so, I realized that I could not continue literature there. That was when I came across critical discourse analysis.
I could immediately see that critical discourse analysis is deeply rooted in the political and literary theory of Marxism, post-structuralism – things we read in literature. Even the term “discourse” is shaped by thinkers like Michel Foucault. I wrote my second MA dissertation on discourse analysis to understand how it works.
My first PhD was also on discourse analysis. I worked on how young Malays talk about global popular culture, especially band music. As they talk about global pop culture, they also talk about themselves. Drawing on Pierre Bourdieu and his idea of taste as a cultural construction, I noticed patterns. Urban students preferred certain radical musicians. Rural students leaned toward others. When I asked about ideals, many eventually referred to local Malay musicians – their own versions of Sabina Yasmin, Runa Laila, or even James in Bangladesh. I also interviewed filmmakers, writers, and intellectuals to understand how they defined Malay identity. Based on all these interviews, I tried to interpret what makes a Malay youth “Malay.”
After that, I taught at the University of Malaya for about six years. I regularly taught discourse analysis and also advanced semantics at the MA level. Through teaching linguistics, I became more engaged with structural aspects of language – phonetics, phonology, semantics, pragmatics, morphosyntax. I do not see sociolinguistics and discourse analysis as separate. Sociolinguists look closely at structure, but both fields examine the relationship between language and society.
Later, I went to Australia and did my second PhD at Macquarie University. I worked on language policy in Malaysia. In 2003, Mahathir Mohamad introduced a policy that mathematics and science should be taught in English. The country was divided over this. I argued in my thesis that education policy is a symptom of nationhood. I used “symptom” in the Lacanian sense. If you only treat the policy, you miss the deeper divisions that produce it.
Studying Malaysia as a Bangladeshi researcher in Australia helped me look at things more objectively.
When I returned to Bangladesh and joined ULAB in 2016, I supervised a couple of students from Rangamati. They were writing about the medium of instruction in classrooms. They showed how students experience cognitive dissonance when they are not allowed to use their mother tongue and must study in a second or third language. Lack of fluency in the dominant language affects performance. That interested me deeply.
Later, at IUB, the Vice Chancellor wanted to establish a center for endangered languages. I was asked to work on it because of my background in language policy. We developed a center focusing on multilingualism, with a wing for endangered languages. I became the Executive Director. My institutional responsibility pushed me further into this field, but the interest was already there.
So, my work in the Chittagong Hill Tracts did not suddenly appear. It grew from my earlier engagement with language policy and divided nations. As a critical discourse analyst, I have always been interested in the gap between state policy and lived experience. The Hill Tracts became a natural extension of that interest.
Interviewer: Your work engages with communities whose languages are often overshadowed by the scenic landscapes of their homeland. How do you negotiate the gap between public fascination with a landscape and the invisibility of those who inhabit it?
Sir: That is a very good and timely question. There are two things here. One is the scenic landscape. The other is the people who inhabit that landscape. The question is: how do the people get erased while the landscape becomes central?
If you look at tourism pages on Facebook, you will see endless pictures of hills, rivers, and clouds. Very little about the people who live there. I have even heard students say, “Chakmas and Marmas are the same.” That tells you something. The dominant population has the means to travel there, to enjoy the landscape whenever they want. But their interest is selective. They want the view. They also want what they call “authentic” Pahari food.
When I was in Australia, I worked on a paper where I asked: when you go to a Pahari restaurant, what are you really consuming? How do you “eat” the hills? The scenic landscape and this desire to consume authenticity are connected. You romanticize the hills in Dhaka. Or you visit a Marma or Mro village with a tourist guide to have an “authentic” experience. You sit with them, eat with them. But the first question often is: is the food halal? So, restaurants respond by offering an “authentic Pahari taste” that is also halal. You get both – romantic differences and cultural comfort.
This is how you keep the landscape but bracket the people. The majority population often wants the hills as part of national identity – “this is my country; this landscape is mine.” But there is less interest in the people themselves.
When we used to take students from IUB to the hills, one of the first things I would tell them was: do not take pictures casually. If you want photographs, take pictures of nature. If you want to photograph a person, ask for permission. The desire for the “authentic Mro face” is not politically innocent. The moment you frame someone as an aesthetic object, it becomes problematic.
You will also notice the absence of scripts. You do not see Marma script or Mro script in public spaces. Even languages like Baum or Pankhua are often written in Roman letters. The visual space does not reflect their linguistic presence.
Interviewer: By not being able to read their stories in their language, we are basically not being able to get the insight, acknowledgement or appreciation of their culture.
Definitely. We are not getting their version of the story. This is about the politics of representation. Films and media have created a romantic image of the hills. We carry that idealized version in our minds. But those lands are inhabited by specific indigenous communities. Who are they? What is their material culture? What is their language like? We are not very interested in these questions.
Many people would not feel comfortable sitting in someone’s house and sharing fried pork belly. Even knowing that such food is cooked in the kitchen can create discomfort. So what do they do? They ask for separate arrangements. Cultural practices – food habits, ways of living – are deeply connected to identity. But it is easier to bracket the people and still enjoy the view.
Interviewer: What aspects of these communities’ linguistic and cultural experiences do you feel deserve more space in public conversations about the region?
Sir: This is a very important question. If I begin with the linguistic aspect, one major issue is that we tend to treat Marma, Chakma, Khumi, Baum, or Pankhua as single, standard languages. But like any language, they have dialects. That dialect mapping has not really been done, or even discussed properly. This is something I am very interested in. I want to work on dialect mapping.
Interviewer: You cannot capture the nuances of a language unless you work on the dialects.
Sir: Precisely. For example, my research assistant, Domiching Marma, is from Khagrachhori. If I show her data from Bandarban, she immediately says, “No, something is wrong with the syntactic structure. This is not how we say it.” So even within the Marma community, there are differences.
If I explain it in Bangla terms, you can say Nodir Marma (River Marma) and Paharer Marma (Land Marma). There are Marma communities who traditionally live near rivers (kyontha) and others who live in the land areas (plaintha). Their dialects differ. Again, the Marma spoken in Bandarban Sadar is different from the Marma spoken in Thanchi, and that again differs from Ruma. So, dialect mapping is crucial. Without it, we miss important linguistic realities.
There is a place called Nafakhum. You must have heard of it. How do you pronounce it?
Interviewer: Nafakhum
Sir: In Bandarban, many places have Marma names, and they also have Bangla versions for outsiders.
Interviewer: Right. Like how Chinese people have both Chinese names and English names.
Sir: Exactly. People want to retain their identity. So, they keep the original name within the community, but for outsiders, a simplified version is used. Now, নাফাখুম (Nafakhum) is actually ঙাফাখুং. Here, ঙা means fish. ফা refers to a specific type of fish – বাঘাই মাছ in Bangla. খুং refers to a particular place in a stream, like an upstream pocket where that fish is found. So ঙাফাখুং means the place where that fish is abundant.
The shift from ঙা to না is linguistic. In Bangla, since the morphosyntactic structure does not allow initial ঙা, it becomes না. That is how ঙাফাখুং becomes নাফাখুম.
I do not mind if a Bangladeshi says Nafakhum. But we need to acknowledge the original form. In official documents, it should be written as ঙাফাখুং.
Interviewer: But that’s the thing, sir. Bangla has such a rich alphabetic system. Even though we don’t usually begin words with ঙা, we can still produce that sound phonologically. For an American speaker, this might be difficult because that sound does not exist in their phonological system. But for us, if we are aware of the correct pronunciation and make the effort, we are fully capable of saying it properly.
Sir: Yes, we can at least acknowledge the actual pronunciation, even if we do not use it daily. Take another example. We were talking about সাঙ্গু (Shangu) river. Do you know its Marma name?
Interviewer: I don’t.
Sir: It is রেগ্রাই খিয়াং (Rigrai Khyong). It means সচ্ছ নদীর পানি – crystal clear river. That meaning disappears when we only use the Bangla name.
I went to Dolupara once. The name comes from Dolubash, a type of bamboo that grows along the riverbank. While working there, I noticed houses with bamboo walls. In the same village, I saw two different weaving patterns. I asked a Marma boy to find out the names of those patterns and the stories behind them.
An elderly woman was sitting there, smoking. She explained that there are four types of bamboo wall patterns. There is even a class distinction regarding who can use which pattern. This is material culture. You see the physical object – the bamboo wall – but you also need the story behind it. When they explain it, they use metaphors and similes. Those rhetorical forms are not decorative; they shape how reality is understood and narrated.
So, when we talk about linguistic and cultural experiences, we must document dialects, original place names, scripts, stories, metaphors, weaving patterns, food practices – all of it together. Language is not separate from material culture. If we want meaningful public conversations about the region, we need to give space to these layered, lived experiences, not just the landscape.
Interviewer: What do you find most compelling about the current directions of language education research in Bangladesh?
departments are already working in this area. They focus on assessment, medium of instruction, and language policy. That is encouraging. Education research is very active at the moment.
But for me, language policy research has always been political. Political in a rights-based sense. It is about what resources people have, and what kind of society they want to build. Questions of equity and social justice are central. When education research touches on these themes, it often introduces them but does not push far enough. It avoids the difficult questions. It does not always ask why inequality exists, or how the divide between haves and have-nots is reproduced.
If I avoid using heavy theoretical terms, I would say that education research in Bangladesh is not yet grounded in asking uncomfortable questions about structural inequality. Take a university like ours. Students come from very different linguistic and socioeconomic backgrounds. Some suggest that we should allow them to speak Bangla in the classroom to make things easier. I find that dangerous. The classroom may be the only space where they can practice and strengthen their English. If we immediately allow them to retreat, we may be limiting their growth.
I am glad that many scholars are working on education research. But I think we need to radicalize it a bit more. In one of my recent works, I wrote about equity and social justice through English medium instruction at the tertiary level, focusing on students from the Chittagong Hill Tracts. When these students, or students from small towns, enter English medium universities, they often lack prior exposure to English. From their experiences, I found that many of them try very hard to turn themselves into what we call “human capital” so they can survive in a neoliberal system. In that system, you are expected to take care of yourself. The institution does not always provide enough structured support.
Even when support systems like English zones exist, many students do not know how to use them effectively. They feel lost. They struggle not only with language fluency but also with content. Because they do not fully command the language, they cannot access the content deeply. So, they suffer at two levels: language and subject matter.
English medium universities have opened new possibilities for education in Bangladesh. But they have also exposed our unpreparedness. We are not fully ready to use a medium of instruction that is not our own. Simply saying that English is a neoliberal language is not enough. We need to show how it functions in our context. How does it reinforce existing inequalities? How does it widen the gap between different groups of students?
Education research needs to listen more carefully to students’ own experiences. We need to document the problems they face in detail. And in the case of the Chittagong Hill Tracts, there is an additional layer. What happens when a student comes with Bangla as a second or third language, and English as a fourth? What kind of cognitive and emotional challenges does that create? That, too, deserves serious research.
Interviewer: Looking ahead, what areas of language research in Bangladesh do you think hold possibilities that we have not yet explored enough?
Sir: When we talk about language research in Bangladesh, we first need to clarify what we mean by “language.” For decades, in many English departments, language has been reduced to ELT. That is problematic. ELT is only one stream. There is core linguistics, applied linguistics, and ELT. These are three different ways of approaching language. Our students need to understand that distinction.
If we take the Chittagong Hill Tracts as an example, Bandarban alone has around 11 to 13 languages. Each of these languages has its own morphosyntactic structure. Field linguistics and language documentation have not been taken seriously in this country. Mostly, students from the linguistics department at University of Dhaka work in this area. Only recently have some private universities begun to support one or two researchers in these fields.
Language documentation should be expanded. When you learn documentation, you learn phonetics, phonology, morphosyntax, semantics, pragmatics, and sociolinguistics. If you know how to analyze phonetic data, you can apply that knowledge elsewhere. A strong foundation in phonetics, phonology, syntax, and semantics increases your chances of securing funding abroad. But we rarely talk about that.
We are planning to offer courses in clinical linguistics and forensic linguistics. Clinical linguistics helps us understand how people speak differently, what phonetic variations exist, how morphosyntactic structures differ, how semantic mapping varies depending on linguistic properties. These areas open new directions for research.
Take semantics. I teach conceptual metaphor theory in my classes. Language is deeply metaphorical. In infancy, warmth is associated with proximity to the mother. Warmth becomes affection. So, when we say someone is “a warm person,” we understand it intuitively. This is where cognitive linguistics comes in. It deals with embodied experience. Language is embodied; it is tied to our physical and sensory experiences. Later, when we enter the symbolic world of language, we draw on those embodied experiences to express abstract ideas.
If students understand cognitive linguistics, they can move into areas like machine translation, large language models, corpus analysis, and sentiment analysis. They can work with large datasets and explore how meaning is constructed at scale. These are areas we have not explored enough in Bangladesh.
So, for me, the future lies in strengthening core linguistics – fieldwork, documentation, phonetics, semantics, cognitive linguistics – and then connecting that foundation to global research areas. That is how we grow and stay connected internationally.
Interviewer: In your opinion, how can younger generations meaningfully engage with Indigenous communities – learning their languages and cultures – in ways that genuinely support, appreciate, and sustain their heritage?
Sir: It is a difficult question. The way it is framed sometimes sounds like older people giving instructions to the young. I am not comfortable with that. In academia, I do not see it as seniors telling juniors what to do. We work together. If a student comes to me and says, “I am interested in this topic,” I expect that we explore it together.
But interest alone is not enough. If you want to work on clinical linguistics, you need to know what aphasia is, what speech delay is. If you want to work on Marma or Mro language, you need to understand language families, morphosyntactic structures. Without that foundation, you cannot meaningfully research these languages.
So, if younger generations want to engage with Indigenous communities, they need to build theoretical grounding first. If you want to document material culture, you should understand linguistic anthropology. You need training. You need to know the conceptual tools. When a student approaches me with a research idea, I expect they have already done their homework. That is part of meaningful engagement. You need to know what the Constitution says about Indigenous rights. You need to understand how these communities differ from the dominant culture and dominant languages.
You do not have to learn every Indigenous language to be empathetic. But you must respect that these are full languages with their own morphosyntactic structures, their own segmental and suprasegmental features. They are not broken versions of Bangla. They operate on different systems.
So meaningful engagement begins with knowledge. Develop theoretical understanding. Understand the epistemic foundations of their language and culture. When you approach communities with that preparation, you are better equipped to support, appreciate, and sustain their heritage in a serious and responsible way.