A diglossic situation exists in a society when it has two distinct codes which show clear functional separation; that is, one code is employed in one set of circumstances and the other in an entirely different set. Ferguson (1959, p. 336) has defined diglossia as follows:
diglossia is a relatively stable language situation in which, in addition to the primary dialects of the language (which may include a standard or regional stand- ards), there is a very divergent, highly codified (often grammatically more complex) superposed variety, the vehicle of a large and respected body of written literature, either of an earlier period or in another speech community, which is learned largely by formal education and is used for most written and formal spoken purposes but is not used by any sector of the community for ordinary conversation.
In the same article he identifies four language situations which show the major characteristics of the diglossic phenomenon: Arabic, Swiss German, Haitian (French and Creole), and Greek. In each situation there is a ‘high’ variety (H) of language and a ‘low’ variety (L). Each variety has its own specialized functions, and each is viewed differently by those who are aware of both.
In the Arabic situation the two varieties are Classical Arabic (H) and the various regional colloquial varieties (L). In Switzerland they are Standard German (H) and Swiss German (L). In Haiti the varieties are Standard French (H) and Haitian Creole (L). In Greece they are the Katharévousa (H) and Dhimotiki, or Demotic (L), varieties of Greek. In each case the two varieties have coexisted for a long period, sometimes, as in the case of Arabic, for many centuries. Con- sequently, the phenomenon of diglossia is not ephemeral in nature; in fact, the opposite is true: it appears to be a persistent social and linguistic phenomenon.
A key defining characteristic of diglossia is that the two varieties are kept quite apart in their functions. One is used in one set of circumstances and the
other in an entirely different set. For example, the H varieties may be used for delivering sermons and formal lectures, especially in a parliament or legislative body, for giving political speeches, for broadcasting the news on radio and television, and for writing poetry, fine literature, and editorials in newspapers.
In contrast, the L varieties may be used in giving instructions to workers in low- prestige occupations or to household servants, in conversation with familiars, in
‘soap operas’ and popular programs on the radio, in captions on political car- toons in newspapers, and in ‘folk literature.’ On occasion, a person may lecture in an H variety but answer questions about its contents or explain parts of it in an L variety so as to ensure understanding.
You do not use an H variety in circumstances calling for an L variety, e.g., for addressing a servant; nor do you usually use an L variety when an H is called for, e.g., for writing a ‘serious’ work of literature. You may indeed do the latter, but it may be a risky endeavor; it is the kind of thing that Chaucer did for the English of his day, and it requires a certain willingness, on the part of both the writer and others, to break away from a diglossic situation by extend- ing the L variety into functions normally associated only with the H. For about three centuries after the Norman Conquest of 1066, English and Norman French coexisted in England in a diglossic situation with Norman French the H variety and English the L. However, gradually the L variety assumed more and more functions associated with the H so that by Chaucer’s time it had become pos- sible to use the L variety for a major literary work.
The H variety is the prestigious, powerful variety; the L variety lacks prestige and power. In fact, there may be so little prestige attached to the L variety that people may even deny that they know it although they may be observed to use it far more frequently than the H variety. Associated with this prestige valuation for the H variety, there is likely to be a strong feeling that the prestige is deserved because the H variety is more beautiful, logical, and expressive than the L variety. That is why it is deemed appropriate for literary use, for religious purposes, and so on. There may also be considerable and widespread resistance to translating certain books into the L variety, e.g., the Qur’an into one or other colloquial varieties of Arabic or the Bible into Haitian Creole or Demotic Greek.
(We should note that even today many speakers of English resist the Bible in any form other than the King James version.)
This last feeling concerning the natural superiority of the H variety is likely to be reinforced by the fact that a considerable body of literature will be found to exist in that variety and almost none in the other. That literature may also be regarded as reflecting essential values about the culture and, when parts of it are classical literature, deemed worthy of recalling by allusion and quotations on occasions suitable for the employment of H. Speakers of Arabic in particular gain prestige from being able to allude to classical sources. The folk literature associated with the L variety will have none of the same prestige; it may interest folklorists and it may be transmuted into an H variety by writers skilled in H, but it is unlikely to be the stuff of which literary histories and traditions are made in its ‘raw’ form.
Another important difference between the H and L varieties is that all chil- dren learn the L variety. Some may concurrently learn the H variety, but many
do not learn it at all; e.g., most Haitians have no knowledge at all of Standard French but all can speak some variety of Haitian Creole, although some, as I have said, may deny that they have this ability. The H variety is also likely to be learned in some kind of formal setting, e.g., in classrooms or as part of a religious or cultural indoctrination. To that extent, the H variety is ‘taught,’
whereas the L variety is ‘learned.’ Teaching requires the availability of grammars, dictionaries, standardized texts, and some widely accepted view about the nature of what is being taught and how it is most effectively to be taught. There are usually no comparable grammars, dictionaries, and standardized texts for the L variety, and any view of that variety is likely to be highly pejorative in nature.
When such grammars and other aids do exist, they have in many cases been written by outsiders, e.g., ‘foreign’ linguists. They are also likely to be neither well known to the people whose linguistic usage they describe nor well received by those people, since such works are unlikely to support some of the myths that accompany diglossia, particularly the myth that the L variety lacks any kind of ‘grammar.’
The L variety often shows a tendency to borrow learned words from the H variety, particularly when speakers try to use the L variety in more formal ways.
The result is a certain admixture of H vocabulary into the L. On other occasions, though, there may be distinctly different pairs of words, i.e., doublets, in the H and L varieties to refer to very common objects and concepts. Since the domains of use of the two varieties do not intersect, there will be an L word for use in L situations and an H word for use in H situations with no possibility of transferring the one to the other. So far as the pronunciation of the two varieties is concerned, the L system will often appear to be the more ‘basic.’ However, actual circumstances can vary. Whereas the two varieties of Greek have very similar sound systems, there is a considerable difference between Classical Arabic and the colloquial varieties and a still greater difference between High German and Swiss German.
Diglossia is a widespread phenomenon in the world, well attested in both space (e.g., varieties of Tamil in the south of India) and time (e.g., Latin in Europe in the Middle Ages). According to Ferguson (1959, p. 338), it is likely to come into being when (1) ‘there is a sizable body of literature in a language closely related to (or even identical with) the natural language of the community . . . [and when (2)] literacy in the community is limited to a small elite, [and]
. . . a suitable period of time, of the order of several centuries, passes from the establishment of (1) and (2).’ People living in a diglossic community do not usually regard diglossia as a ‘problem.’ It becomes a problem only when there is a growth of literacy, or when there is a desire to decrease regional and/or social barriers, or when a need is seen for a unified ‘national’ language.
In Haiti, any attempt to develop literacy had to confront directly the issue of whether to increase the amount of Standard French taught or to ‘elevate’ the L variety, Haitian Creole, into a national language. Haitian Creole was eventu- ally recognized as a national language in 1983, with prestigious French, of course, the other. Both languages were made official in 1987. There has been an ongoing debate about the most appropriate orthography (spelling system) for Haitian Creole: about the use of certain letters and accents, and about whether the differences between French and Haitian Creole should be minimized in the
orthography for Haitian Creole or whether that orthography should be as trans- parent as possible in relating letters to sounds, particularly the sounds of the most widespread variety of Haitian Creole. French, though not widely used, has such prestige that, according to Schieffelin and Doucet (1998, p. 306) virtually any proposal for an othography for kreyòl has created ‘resistance both to the adoption of the orthography and to the use of kreyòl as a medium of instruction in school. The double resistance comes from both the masses and the educated elite minority. The masses see the officialization of written and spoken kreyòl in school as limiting their access to French and, consequently, their social and economic mobility. The elites, who already know kreyòl, do not see the point of teaching it, in any form, in school.’
The Greeks have still not entirely solved the problems associated with their two varieties: ‘conservative’ Greeks want to resolve any differences in favor of the H variety, but ‘liberals’ favor the L variety. (It was at one time said that you could judge a Greek’s social and political attitudes by the way he or she declined third-declension nouns!) The twentieth century witnessed a long and sometimes bitter struggle between supporters of the two varieties. Religious authorities condemned a 1921 translation of the New Testament into Demotic Greek and this action led to rioting in the streets of Athens. One consequence of the language disagreement was that, when the ‘liberal’ government of the 1960s was overthrown by the ‘colonels’ in 1967, the former government’s program to extend the uses of Dhimotiki was superseded by restoration of use of the H variety, Katharévousa, for example in education, and the suppression of Dhimotiki because of its association with ‘left-wing’ views. With the return to constitutional government in 1975 the H was superseded in turn by the L, Dhimotiki was declared the official language of Greece in 1976, and Katharévousa disappeared almost entirely from public view. The new model for Greece seems to be based on the variety spoken in Athens. Today, the opponents of this new Greek language based on the L variety attack it for being impoverished and cut off from its roots, which are said to be the former H variety and Ancient Greek (Frangoudaki, 1992). Tseronis (2002) says that the two most recent Greek dictionaries, the Dictionary of Modern Greek Language (DOMGL) and the Dictionary of Common Modern Greek (DOCMG) show that the process of standardization continues. The DOMGL finds its roots in Katharévousa and the DOCMG in Dhimotiki. However, both point to eventual unification around the variety spoken in Athens and an end to the H–L division.
The linguistic situations in Haiti and Greece are intimately tied to power relationships among social groups. Traditionally, in each country the H variety has been associated with an elite and the L variety with everyone else. Diglossia reinforces social distinctions. It is used to assert social position and to keep people in their place, particularly those at the lower end of the social hierarchy.
Any move to extend the L variety, even in the case of Haiti to make the population literate in any variety, is likely to be perceived to be a direct threat to those who want to maintain traditional relationships and the existing power structure.
The following example from Trudgill (1995, pp. 101–2) shows how different the Zürich variety of Swiss German is from High German:
Low variety – Swiss German
En Schwyzer isch er zwaar nie woorde, weder en papiirige na äine im Hëërz ine;
und eebigs häd mer syner Spraach aagm2rkt, das er nüd daa uufgwachsen ischt.
Nüd nu s Muul häd Ussländer verraate, au syni Möödteli. Er häd lieber mit syne tüütsche Landslüüte weder mit de Yhäimische vercheert, und ischt Mitgliid und Zaalmäischter von irem Veräin gsy.
High variety – Standard German
Ein Schweizer ist er zwar nie geworden, weder auf dem Papier noch im Herzen;
und man hat es seine Sprache angemerkt, dass er nicht dort aufgewachsen ist.
Nicht nur die Sprache hat den Ausländer verraten, sondern auch seine Gewohnheiten.
Er hat lieber mit seinen deutschen Landsleuten als mit den Einheimischen verkehrt, und ist Mitglied und Zahlmeister ihres Vereins gewesen.
English
He never actually became Swiss, neither on paper nor in his heart; and you could tell from his language that he had not grown up there. It was not only his language that showed that he was a foreigner – his way of life showed it too. He preferred to associate with his German compatriots rather than with the natives, and was a member and the treasurer of their society.
Swiss German diglossia has its own stabilizing factors. Switzerland is a multi- lingual country, with German, French, and Italian its three official languages.
Strong constitutional protection is provided for German, the H variety of which is taught in the schools and used in official publications, newspapers, literature, and church services. This allows the German Swiss to communicate with speakers of German elsewhere in Europe and gives them access to everything written in Standard German. However, the Germans in Switzerland can also assert their independence of other Germans through use of their L variety. This is their own distinctive unifying spoken variety of German, one in which they take a special pride. The continuation of the High German–Swiss diglossic situation depends every much on the continued effectiveness of educating Swiss German children to use High German in the schools so as to encourage diglossia there. Some Swiss do worry that such teaching of High German may not always produce the desired results and that any quest for identity through increased use of Swiss German might lead to growing cultural isolation from other users of German.
In much the same way, the people of Luxembourg have achieved a certain distinctiveness with their own diglossic – or better still, triglossic – situation (see Newton, 1996). In this case Luxemburgish, called Lëtzebuergesch, a variety of German, is the L variety and Standard German is the H variety. The following examples are from Trudgill (1995, p. 103):
Luxemburgish
Wéi de Rodange 1872 säi Buch drécke gelooss huet, du bluf hien drop sëtzen. En hat e puer Leit ze luusség op d’Zéiwe getrëppelt, déi dat net verquësst hun. Eréischt eng Generation doerno huet de Rodange uge-faang séng giedléch Plaz ze kréien.
Séng Kanner hu wéinstens nach erlieft, wéi 1927 eng Grimmel vun deem gutt gemaach guf, wat un him verbrach gi wor!
Standard German
Als Rodange 1872 sein Buch drucken liess, hatte er keinen Erfolg damit. Mit zuviel List war er ein paar Leuten auf die Zehen getreten, und die konnten ihm das nicht verzeihen. Erst eine Generation später begann Rodange, seinen ihm zustehenden Platz zu erhalten. Seine Kinder haben es wenigstens noch erlebt, dass 1927 ein wenig von dem gut gemacht wurde, was an ihm verbrochen worden war!
English
When Rodange had his book printed in 1872 he had no success with it. With too much intrigue he had trodden on some people’s toes, and they could not forgive him that. Only a generation later did Rodange begin to receive his rightful place.
His children at least experienced the making good, in 1927, of some of the wrong that had been done him.
However, the situation is a little more complicated in Luxembourg than in Switzerland because still another language, French, is involved. All three languages – German, French, and Luxemburgish – have been official languages since 1984.
Inhabitants of Luxembourg not only use Luxemburgish (e.g., in ordinary con- versation) and Standard German (e.g., in letter writing, books, and newspapers), but they also use French (e.g., in parliament and higher education) – see Clyne (1984, pp. 20–1). Moreover, they frequently borrow words from French for use in Luxemburgish. Consequently, it is not unusual for a speaker of Standard German who goes to live in Luxembourg to feel that Luxemburgish is a variety of French rather than a variety of German! French is highly regarded in Luxem- bourg and is also the most widely used language (by 96 percent of residents), although 81 percent can speak German and 80 percent can speak Luxemburgish (Fehlen, 2002, p. 91). However, the clear marker of Luxembourg identity among Luxembourgers is their use of Luxemburgish; it is a solidarity marker just as is the use of Swiss German among Swiss Germans.
The Arabic situation is very different again. There are a number of flourishing regional varieties of the L and many Arabs would like to see the Arab-speaking world unify around one variety. They acknowledge the highly restricted uses of the H variety, but also revere it for certain characteristics that they ascribe to it: its beauty, logic, and richness. Classical Arabic is also the language of the Qur’an. Ferguson has pointed out that choosing one colloquial variety of Arabic to elevate above all others poses a number of problems. Almost certainly, any Arab will tell you that the variety he or she speaks is the ‘best,’ so there would be considerable disagreement about where one should begin any attempt to standardize modern Arabic on a single colloquial variety. There is, however, a consensus among Arabs that any standard that may eventually emerge will be a version of the H variety developed to meet modern needs and purged of regional peculiarities and foreign impurities.
While acknowledging that diglossic situations are essentially stable, Ferguson did predict (1959, p. 340) what he thought the future held for the situations he
examined. He regarded the situation in Switzerland as relatively stable. The Arabic one seemed to point to the development of several regional standard varieties of Arabic, each using a considerable amount of vocabulary drawn from Classical Arabic. In Haiti, there would be a slow development of Haitian Creole based on the L variety of the capital, Port-au-Prince. Finally, in Greece the standard would be based on the L variety of Athens with considerable admixture of vocabulary from Katharévousa.
What Ferguson describes are ‘narrow’ or ‘classic’ diglossic situations. They require the use of very divergent varieties of the same language and there are few good examples. Fishman has broadened or extended the term to include a wider variety of language situations. For Fishman (1980, p. 3) diglossia is ‘an enduring societal arrangement, extending at least beyond a three generation period, such that two “languages” each have their secure, phenomenologically legitimate and widely implemented functions.’ By acknowledging that his use of the term language also includes sub-varieties of one language, Fishman includes Ferguson’s examples. He does add, though, that in the case of two varieties of the same language, they be ‘sufficiently different from one another that, without schooling, the elevated variety cannot be understood by speakers of the vernacular’ (p. 4).
Fishman’s proposal extends the concept of ‘diglossia’ to include bilingual and multilingual situations in which the different languages have quite different functions. For example, one language is used in one set of circumstances and the other in an entirely different set and such difference is felt to be normal and proper. Fishman gives examples such as Biblical Hebrew and Yiddish for many Jews, Spanish and Guaraní in Paraguay, and even Standard English and Carib- bean Creole. I will have more to say about code choices in such situations in the following sections.
Discussion
1. In what ways was the relationship between Classical Latin and the vernacu- lar Romance languages, particularly the languages presently known as French, Italian, Spanish, and Portuguese, a diglossic one for a number of centuries?
At which point did diglossia cease? Can we answer this last question with any great degree of precision?
2. The history of English in the three centuries after 1066 is of considerable interest. The Norman Conquest established Norman French as the H vari- ety and English as the L variety. What caused English eventually to triumph and French to be eliminated from use in England (except as a clearly marked
‘foreign’ language)? Where did Latin fit into the overall picture?
3. There was, and still is, among certain ultra-orthodox sects in Israel resist- ance to the use of Biblical Hebrew as a vernacular language, that is, as a language of everyday living, because they regard such use as ‘profane.’ How different are Biblical and Modern Hebrew?
4. If someone were to tell you that diglossia is but a simple reflection of the social, cultural, or political oppression of a people, how might you answer?