Professor Gérard Diffloth is a leading figure in Southeast Asian linguistics, specializing in the languages of the Austroasiatic family that includes Khmer, Vietnamese and many other languages spoken not only in the countries of Southeast Asia, but also northeast India, Southern China and the Nicobar Islands. His main work has been concerned with elaborating the linguistic history of the region. He and Nathan Badenoch are working on a group of small and endangered languages spoken in northern Laos. This interview arises out of an exchange on this project and other work related to it during Prof Diffloth’s recent stay as a Visiting Scholar at CSEAS.



Badenoch: First, I know that you have been here at the Center for Southeast Asian Studies (CSEAS) once before. When was that and what were you doing then?

Diffloth: I was at CSEAS in 1976–1977. The Center was much smaller then, yet it was already a very important place for Southeast Asian studies. At that time I was working with ­Yasuyuki Mitani (三谷恭之), who was finishing his dissertation on the Lawa language of Northern Thailand, so we were able to discuss common interests. Professor Ishii Yoneo (石井米雄 1929–2010) was also here at the time and he was the first ­person to teach me Thai.

Badenoch: You were already well into your career as a linguist by then, but you had initially studied mathematics and journalism. How did you end up as a specialist of Southeast Asian linguistics?

Diffloth: That is a very long story. But I should say that it was really by chance, or actually luck, that I moved on to Southeast Asian, and especially Austroasiatic linguistics. Along the way there was mathematics at the University of Paris and journalism at the Higher School of Journalism of Lille, also in France. At that school, I started studying Russian, and this was an eye-opener. The Russian language in itself is very interesting, and the textbooks used for teaching Russian were quite sophisticated in linguistic terms. That is when I understood that there was such a thing as the scientific study of language, and my background in mathematics became useful then.

Badenoch: I recall in his book The Road will Open (Michi wa Hirakeru­) how Professor Ishii described making several attempts at learning Russian. He did not make anything of it because he did not find inspiration in that language. Clearly you had a different experience. Nonetheless, you moved from Russian to languages further east. How did that happen?

Diffloth: Gradually, via Persian and then Tamil — I wrote my dissertation on a variety of Tamil called Irula — and from Tamil to the mountains of Malaysia. And from there to the rest of the Austroasiatic family across the whole of Mainland Southeast Asia. But in a more personal way, by interacting with speakers of languages such as Khasi, Semai, Mon, Kuay, Khmer and others, I found that the Austroasiatic family was unique in many ways, and historically very rich.

Badenoch: Of course linguistics is a very broad field, but I feel that there are many misconceptions or misunderstandings about what linguists do. There are two misconceptions. Firstly, that linguists are general polyglots; and secondly, that linguistics is impenetrably technical, inaccessible to non-specialists, and unfortunately, not relevant for other fields of research.

Diffloth: You are right in saying that linguistics is often misunderstood. It’s a great pity, because when you think about it, and this is something I first learned in journalism school, language is with us everywhere all the time, from the moment we wake up until when we fall asleep, and often in our dreams as well. The language we speak is like the air we breathe, obvious and yet invisible. And Linguistics has developed scientific concepts and tools that are powerful and specifically designed for this very human activity, language.

Badenoch: As you made the transition into linguistics, what people were influential on your thinking?

Diffloth: It really goes back to early years, even to pre-school times. Thinking about it, I find that some teachers, even in those early days, had a great influence. Primary school teachers have a very important job to do. Young children have a brain, of course, just like adults. An inspiring teacher, a casual remark, a convincing example, can create a spark that may keep on shining for many years.

Badenoch: Children’s attitudes towards language are set at a very young age. Unfortunately this is often in a negative way, as we can see in the way English and other languages are taught around the world. Were you surrounded by people speaking other languages when you were young?

Diffloth: Not that much; but at the age of four, in the countryside during the war, I very well remember being astonished at hearing some old farming ladies speaking, not my standard French, but a variety of Berrichon. Later on in school, Latin was vigorously taught early on, and then Greek. English was taught as a secondary language. My father also loved to travel and we often visited countries where everybody spoke German. So there was a mixture of the classics and the spoken languages of Europe.

Badenoch: One of the things that has underpinned your linguistic career has been fieldwork. What do you think the ­experience of doing intensive fieldwork can teach us about language?

Diffloth: Very quickly you realize that language is not an object, but an activity. Unfortunately, in societies where literacy is well implanted, most people think, — because they are taught this in school —, that language is basically writing, the sort of black stuff you can see on paper as text. But written text is not language, and we have been made to forget that this writing is actually derived from language, not the other way around. One of the first things you witness in doing fieldwork is that language is something people do, not something people make. There are still societies today where the idea of putting down language on paper appears quite senseless, even objectionable.

Badenoch: When doing fieldwork have you felt any tension between Western academic science, which is strongly based in written culture, and the language that people are speaking?

Diffloth: Yes, but when you do fieldwork for a long time, you begin to see things the way they do. To give an example, at some point in studying the Mon-Khmer languages of Malaysia, I was going through a certain type of words — Expressives, somewhat similar to the Gisego (擬声語) found in Japanese — with a native speaker of Semai. At some point he said to me: “Actually, these words which you call Expressives, they are not really words at all. Up until now, we have been discussing nouns, verbs, and so on, and that is all very fine, but these things are different: we do not speak them, we actually shoot them.” I struggled to understand what he could possibly mean by that; and it has taken me some years to draw the linguistic conclusions from his strange remark.

Badenoch: A German linguist named Rudi Keller has described language as being an agreement between speakers. This agreement is what determines meaning, changes in sound and word use over time (Keller 1994). Perhaps there is a certain type of ‘agreement’ that we work towards with the people whose language­ we are studying in the field?

Diffloth: Absolutely. When we are notating the languages we hear, using whatever notation system, these notations are not language. They are simply the permanent record of an agreement between you, the linguist, and the speaker: we have come to an agreement, viva voce, that this is the way it sounds, and then the linguist makes a note of that agreement. This notion has direct consequences when it comes to recording dying languages, a major concern in linguistics today. Actually, the term “recording” is ambiguous. In its superficial sense, it ­implies that we take along some high-tech equipment, make hundreds of hours of excellent “recordings” and then go home to study them, or store them for the future. But unless you have been constantly interacting, on a word-by-word basis, with the very person who spoke, who told the stories, who sang the songs, such recordings are practically worthless. Recordings, in the deeper sense of the term, are witnesses of detailed and elaborate agreements.

Badenoch: Speaking of language loss, Southeast Asia is known as an area of great linguistic diversity, but the rate at which languages are disappearing is very high. Of course this reflects a global trend, but in Southeast Asia why are languages disappearing so quickly?

Diffloth: The figures are staggering — in 50 years, half of the languages spoken in the world today may well have disappeared forever. If you compare that situation with the current threat of extinction of plant and animal species worldwide, the ongoing rate of language elimination is much more dramatic; and yet, it is not making many headlines. In 1988, the United Nations celebrated a “year of the languages”; Kofi Anan’s recommendation­ was to “protect and promote” endangered languages­. Since then, there have been a few projects aimed at promotion; but when it comes to protection, I don’t see that much has been done, at least in Mainland Southeast Asia. The main reasons appear quite simple: having many languages and dialects makes the work of administrators difficult and costly; also, a certain model of the nation — the monolingual nation — seems dominant, forming a considerable obstacle. Following that model, the national languages are being well protected and forcefully promoted.

Badenoch: You mentioned the double-pronged approach of promoting and protecting. Protection implies that there is a threat, but I don’t believe that people have yet recognized that the threats have been identified.

Diffloth: I agree. Even the words we use in discussing the problem: this language is ‘dying’ or ‘moribund,’ or ‘comatose,’ all of these medical metaphors, imply that like living organisms, languages must have a birth and a death. These metaphors are totally misleading and confuse the debate. The fundamental dynamics of language, of any language, consists in projecting itself indefinitely into the future, linking one generation to the next. In that context, the elimination of a language is always the result of multiple conscious decisions taken at many levels. The reasons why people decide to speak one language or another, either in public, or to their friends, or in their own house, and most importantly to their own children, are publicly complex, and mostly intimate.

Badenoch: We are constantly faced with the enormous question, what is lost when a language disappears?

Diffloth: I recently gave a talk at CSEAS where I said that the loss of a language is like the sinking of a cathedral. The idea was that languages transport very complex and not always obvious­ ideas, and ways of perceiving practically every aspect of life. In addition to what people openly say — the narratives of culture, traditions and ways of life — the languages themselves contain rich information that is not overtly stated but only implied and yet well understood by all. This is the result of hundreds, thousands of years of experience and observations, of success and tragedy.

Badenoch: The experience of humanity?

Diffloth: Yes, this is History with a capital H, and different languages­ are the result of different histories. In this sense, when a language disappears, it is as if a cathedral collapsed or a library was burnt to the ground.

Badenoch: You have often said that language is history. Can you elaborate a little more on what insights can be obtained from this language perspective on history?

Diffloth: When we get involved in historical linguistics, the results very often end up being rather different from the histories produced from the analysis of concrete historical documents. For one thing, traditional history usually has to do with power structures, governance, armies and battles, things of that kind; historical linguistics can do this as well, but also gets into the minutiae of life: the history of dress, of hunting, of family arrangements. Another difference is that in historical linguistics we are compelled to look at minority languages because they are useful, and historically every bit as legitimate as the major, the usually written national languages. Quite often, the histories of people without writing are simply absent from the more traditional narratives. Sometimes, what we find with the use of historical linguistics squarely contradicts what is said in the history books.

Badenoch: Oral history has become more and more a part of social science, but you are talking about a different kind of orality­. Not only the stories, but the languages themselves.

Diffloth: Exactly. Oral history and village history, based on living­ peoples’ memories, are without any doubt interesting and important. But the history that we can produce by using historical linguistic methods is of a different kind. Very often we can reach very much further back in time, often discovering events that have left no conscious traces in living memories, or may even have been carefully and completely erased. The reason is that people are not aware of linguistic change. With historical linguistics we are exploring a kind of pre-conscious domain of history.

Badenoch: One type of finding we often talk about falls into the category of food history. The vocabulary of hunting, gathering, and processing different foods is incredibly rich in the Austroasiatic languages.

Diffloth: Every word, each with its own meanings, has a history that we can explore; for example the history of food collection and preparation, the history of culinary tastes. This is something we can often do quite well, given sufficient data. It will soon be possible, for example, to trace the history of when and how rice became a staple food, and what the position and uses of rice may have been before that. This subject has now become a major topic in archeological research.

Badenoch: And I would say this is a prime example of the heritage that is lost when a language ceases to be spoken. The situation is quite bleak, but what do you think should be done?

Diffloth: The question should be “What can we do in the time available to us?” It is urgent. I have seen projects aiming at language re-vitalization, and I wish them the best of luck. But in many places, there is no time left for this type of work. So we are faced with the necessity of doing salvage linguistics. The boat is sinking and we need to save whatever we can. For some of us, this means recording vocabulary. For others it means recording syntax, stories, songs; everything is valuable.

Badenoch: In my research I study how language use changes in situations where many different ethnic groups are living together­ in rapidly growing villages in rural Laos. I see the negative influence that education policy has on minority languages­. Do you think that education is an area that can help improve the situation?


Diffloth: The main problem is that a great number of the languages­ that will disappear are not, and have never been written. This means that an enlightened education system that would allow the use of different languages in the development and delivery of the official curriculum is simply outside the scope of the problem at hand. If a language is no longer spoken­, that means it has already disappeared, it is essentially too late.

Badenoch: Hence the need for salvage linguistics right? In the countries of Southeast Asia, linguistics is a relatively new field of study. Since much of our work at the Center is done in collaborative arrangements with local institutions, I wonder what you think of the prospects for raising the profile of linguistics in local academic research.

Diffloth: Linguists is indeed a new field, although an interest in language is quite old. Until very recently it was mostly expressed in the framework of philology, the study of texts and inscriptions, in Cambodia for example. But linguistics is a very different enterprise. And you are right in saying that linguistics remains a minor subject of study in Southeast Asia, with meager financing and hardly any job prospects . It will take time.

Badenoch: And this difficulty is certainly not limited to Cambodia­.

Diffloth: No, in fact it may be even more difficult in some other countries. One problem is that the collection of primary linguistic data has sometimes become entangled and confused with religious and ethnic-identity issues.

Badenoch: Fieldwork in local languages is at the core of Area Studies at Kyoto University. However, it is my impression that the study of language itself had a somewhat higher profile at the Center when you were here the first time.

Diffloth: I’m not sure; I wouldn’t say that the position of language has fallen all that much. The Center has always placed high emphasis on learning and using local languages, and this is one of the strong parts of the program at the Center. There are several dimensions to this. First of all, if we are going to do any kind of fieldwork, this cannot be done in a short period of time. A long-term, in situ commitment is required. Another side of this issue is what we mean by ‘local’ language. Fieldwork may range from work with officials conducted in the national language, all the way to situations in which the national language is not the language of everyday communication. At this end of the spectrum, we are dealing with languages that are very often not taught anywhere, and in many cases may not have ever been studied. This wide array of possibilities must be kept in mind when we discuss the principle of working in local languages.

Badenoch: So even if one is not a linguist, there is a chance that a researcher might be the first person to work in a minority language.

Diffloth: Yes, and this brings us back to our original discussion; language is everywhere at all times of day and night. Studying linguistics and its principles makes you aware of the multiple dimensions of language, such as phonetics, intonation, syntax, social interaction, semantics and more. These are relevant everywhere, in all aspects of research and well beyond linguistics.

Badenoch: I wouldn’t suggest that everyone should be trained as a linguist, but given the commitment to fieldwork, it seems reasonable that some linguistic training be part of the basic skill-set assumed of a field worker.

Diffloth: This is useful for the researcher, and it is also very important from the perspective of the community where research is being conducted. Because what language you speak, and how you speak it, will give the other persons a certain image of who you are. In the case of a minority language, when you reach a level where you begin to sound like a normal speaker, you suddenly become a fuller human being in their eyes, maybe a member of the household — as opposed to someone who is using only the national language, or is content with being roughly and minimally understood. A speaker of Semai, one of the Mon-Khmer languages in Malaysia, once said to me “It is funny, when I speak Malay and when I speak Semai, the truth is not the same.” A different truth! It’s not simply that we use different words or talk about different topics; what is true in one language may not be true in another.

Badenoch: Finally, having explored these truths through many different languages over the past four decades, how would you describe the ‘project’ of your career in Austroasiatic linguistics?

Diffloth: The general idea has long been to compare languages and reconstruct the previous state of affairs many centuries ago, and as far back in time as the method permits to go. Sometimes, this may require paying close attention to some very fine details of vowels, of tones, and of consonants. This is why linguistics is sometimes seen as a dry and technical subject. But working in this way also provides a guarantee that we can produce testable hypotheses, not just gratuitous fantasies. Such precision is costly in time and labor, and this is true in any field of science. For me, the methods and the tools are fascinating all by themselves; but there is also the much larger goal of understanding meaning, and changes in meaning, and how these changes reflect the history of human experience in a part of the world that is as little known as it is genuinely welcoming.


Ishii Yoneo 石井米雄. 2003. Michi ha Hirakeru: Tai Kenkyu no 50 Nen. 「道は、ひらける――タイ研究の50年」 [The Road will Open]. Tokyo: Mekong Publishing.

Keller, Rudi. 1994. On Linguistic Change: The Invisible Hand in Language. New York: Routledge.