Material Type



Media is loading


A talk by Anthony Pym in a course in variation in English. - Transcript below:
What is diglossia? It’s from Greek: di- means two; gloss, the tongue. Two languages. Not to be confused, however, with bilingualism, which is from Latin: bi-, two: lingua, the tongue. Two languages.

There is, however, in English social linguistics a systematic difference between the two terms, diglossia and bilingualism. Usually, bilingualism is the capacity of the individual, of a person, to speak one, two, or three—more than one—language, let’s say: bilingualism, okay? You could call them polyglots, that’s a nice term for describing people, and French and French-inspired social linguistics talks about plurilingualism for the capacity of the individual. Now, diglossia is something quite different. Diglossia is a social situation; it’s not concerning individuals, it concerns a society in which there are two languages related in such a way that they have different social functions. Okay? That’s diglossia: a social situation; bilingualism, plurilingualism is concerned with the capacities of the individual. Now, a standard definition of diglossia—this is [Charles] Ferguson, 1959—oh, it’s long and complicated, but anyway, diglossia is a relatively stable language situation. And that’s important; it’s not a transitory thing, it’s not a bad thing, it’s something that we observe occurring over centuries in many parts of the world. So, a situation in which, in addition to the primary dialects of the language, there is a very divergent, highly codified (often grammatically more complex) superposed variety. So we have these two kinds of varieties happening within the same language; one would be spoken—the dialects, etc.—and the other would be learned, standardized, the language of literature. Then he goes on of written literature either of an earlier period or in another speech community, which is learned largely by formal education—so you get to this other one by going to school—and is used for written and formal spoken purposes but is not used by any section of the community for ordinary conversation. So it’s easier to understand if you go to Zurich, for example, where you’ve got people speaking Swiss German in the street and on television, on local television, and then going and studying in standard German and learning to write standard German, and they wouldn’t write down their spoken language. These two varieties of the language with different social functions, and they are highly separate. Another classic example would be Arabic in Morocco, where we do have classical Arabic for religious functions, certainly for the King, and then spoken Moroccan Arabic in the street, although Moroccan Arabic does get into the press in that case, okay, So those are cases where the one language has varieties with different social functions. The functions are traditionally called H and L in English. H stands for high, but you don’t say high; H stands for the written, official social functions. L stands for the spoken, non-official, vernacular social functions; low, okay. We try to avoid high and low because that was Charles Darwin’s mistake, when he talked about the higher species, that led to all sorts of racism and misunderstandings. H and L are there not in the sense of H being superior but of them simply being different. That’s why the decision has been made to use H and L as letters rather than as descriptors.

Now that’s a strict definition of diglossia. There’s a more relaxed definition, and that would be when the two varieties in question don’t have to belong to the same language, okay? So in parts of the complex society around us here, we find Spanish being used for official functions. Certainly, here, 50 years ago, Spanish would be absolutely the H variety and Catalan would be the L variety. They are different languages—cognate, but different—and yet they would satisfy most non-demanding definitions of diglossia. So that would be the relaxed definition, or the loose definition: the two varieties, two different functions. The varieties don’t have to belong to the same language; they can, but they don’t have to, okay. I’ll point out that now with the standardization of Catalan—so it’s become very much the H variety around us here—we find situations where Catalan occupies H functions in official society, certainly Barcelona. Spanish can move to L for many of the immigrant groups and occupy those functions, and then we have another Catalan, which is that of the farmers and the traditional working class, with its many regional varieties, and that’s becoming an L as well. So it needn’t be just H and L. There can be other languages, or the same language can move into those two positions if, uh, if the society takes on that sort of form. Um, when we— when we use— Catalan linguists don’t like the theory of diglossia and the basic reason is this: diglossia sort of accepts asymmetries; it accepts that language is going to have different power relations, and that this is a stable and normal thing. Whereas their fight has long been for Catalan to assume full H functions, and the official language policy in Spain is for all co-official languages to have full H functions. So they want a situation that they call bilingüisme, which is H and H full capacity in everything. Why not? That can happen; there’s no law against it. The simple observation in English-language social linguistics is that it needn’t happen, that we have long-term stable asymmetries in language functions. So, if you find that you haven’t got it, it’s not because you’re an aberration, it’s just because your societies tend to suggest that we can have asymmetric language functions without any disaster befalling anybody. The other thing that, um, that my students will say is that “we don’t want our language to have an L function—L means powerless; H means power. Give me power, empower me, make my language big and strong and written and standardized.” Which, of course, is what any linguist would do because linguists are the people who do that sort of work. Great work for ourselves, yeah. All right, but be careful. Over history, the languages that die are often those that are in the H position. Look no further than classical Greek or Latin. All the romance languages that we speak had an L function in relation to H-variety Latin. Which one won out over history? The L varieties, not the H. English itself is the result of a diglossic situation where we had Old French in H we had Anglo-Saxon varieties in L. And did H repress L and kill L over time? Quite the opposite. The result, the English that we have is a merger of the two but with a

rising influence, I suspect over time, of the L. The L came up and absorbed the H. So it’s not true that it’s bad, historically, to be in an L position. An L position is close to where the people are and economic activity is and where people vote, after all. In our course we look, of course, at certain things that depend on diglossia. Diglossia is like the basic social situation that sets up the possibility of, for example, a lot of code-switching that we find. And then if you think of the example of Oberwart where, uh, Hungarian and German were in contact we found that the language shift that we saw there was a classic case of what we now know and would call diglossia, where German had the official function, the H functions, Hungarian had the social life, the association with territory over time. And in that particular case, because of the political shift of the village, the H took over and displaced L in that particular situation. There are no fatalities. It’s not always bad to be in the L position, and H and L relations in diglossia can continue and be stable for many centuries. That’s the lesson, at least, of English social linguistics. You’re welcome to find counter-examples.

Media Format




Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

OER Category

UMSL OER Adopted

Publication Date


This OER at UMSL

Prof. Allard - Foreign Language 2100

This document is currently not available here.