...Pennsylvania

As submitted to
The Handbook of Sociolinguistics, Florian Coulmas (ed.), Basil Blackwell, Ltd., 1997

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...(1959).

The phenomenon was mentioned earlier, as diglossie, in the work of Marçais (1932-33).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...Europe.

There is neither time nor space here to review the literature; the reader is referred to Ferguson's pioneering article (Ferguson 1959), Fishman's extension of diglossia to non-genetic situations (Fishman 1967) and some of the more recent literature (and controversy) on the problems of typologizing diglossia (such as Britto 1986.) Recent state-of-the-art studies are Hudson (1991a) and the bibliographies in Hudson 1992 and Fernandez 1993.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...genres.

In Tamil, the conversational portions of novels and short stories are in L-variety, but not the narrative or descriptive portions.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...tatsamas,

Sanskrit for words borrowed `as is', without phonological adaptation in the host language.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...conceivable.

In South Asia, there is evidence that a highly-structured oral system for the transmission of sacred texts was, and to some extent still is, in place. The reliance on orality was motivated in part by the power of spoken words to invoke the intervention of the gods. In the Indic tradition, if the text has been learned in the proper way, and by the proper person (only male members of the priestly caste may receive the long training involved in the learning, by rote, of the texts) then the power of the word, when spoken, is irrevocable--the gods must act, and will act. Writing the word on paper (stone, copper, whatever) is thus not a substitute for pronouncing it. The utterance of an invocation is thus automatically what modern speech-act theorists would call a performative speech act. In the saying of the word, something is also done, and cannot be undone. The mode of transmission, orality, involved memorization beginning at a young age, and the willingness to devote great amounts of the society's labor and resources to achieve the goal of maintenance and transmission of this textual tradition. Having set this in motion, it also became a cultural value to preserve the infrastructure needed to propel the system--a system of gurus, pandits, disciples, and in some cases monasticism; and of course the caste system with a special niche and privileges for the (hereditary) priesthood.

Though diglossia in most Indian languages now involves literacy, there are still pre-literate (or only orally literate) groups like the Todas of the Nilgiri Hills, for whom we can discern three distinct varieties of speech: spoken Toda, sung Toda (not automatically comprehensible to someone knowing spoken Toda) and trance-language Toda (probably a pidginized kind of Malayalam, but as yet unresearched) which can only be spoken by people in a trance, but can be understood by initiates who are in an unaltered state of consciousness. The central position of Toda songs in Toda culture has been documented thoroughly by Emeneau (1964, 1974); the body of songs is their only form of literature, and can be seen to be in a kind of diglossic relationship to spoken Toda.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...languages

Fishman later elaborated on his 1967 departure from the original presentation of Ferguson, who limited diglossia to ``...speech communities [where] two or more varieties of the same language are used by speakers under different conditions." with `` no attempt ...made ...to examine the analogous situation where two distinct [related or unrelated] languages are used side by side throughout a speech community, each with a clearly defined role." (Ferguson, 1959: 325). For Fishman's 1980 taxonomy, see below. He notes at least four different kinds of diglossia, but indicates that there are of course ``various more complex cases within each of the ...major clusters ...." including cases with more than one H variety, or more than one L-variety. (Fishman, 1980: 4)

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...etc.

Modern sinological linguistic usage for these terms are Classical Chinese and Putonghua or Beijing dialect of standard Chinese.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...4).

To distinguish the different kinds of genetic relationships in (a) and (d), let us use the analogy of consanguineal descent for (a), that is L is descended from H, i.e. is a daughter of H, whereas in (d), the two are descended from a common ancestor, but laterally, i.e. as distant cousins.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...define.

Britto also provides a useful annotated glossary of linguistic terms (1986: 295-333), as well as a review of some of the controversy, if that is not too strong a word for it, surrounding Fishman's extension of Ferguson 1959.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...culture

By this is meant the set of behaviors, beliefs, myths, attitudes, and historical circumstances associated with a particular language.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...silent

Especially if it is a case of Fishman's type (d), where there is a written/formal-spoken norm.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...unnatural.

Even in America one sees a style shift in these same genres of broadcasting, for example when an anchorperson finishes reading a prepared news story and turns to someone in the field for an on-the-spot report, or at least a more relaxed discussion of something: ``We're gonna go now to Tom Brokaw, who's on the floor of the Convention...".

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...HREF="footnode.html#156">

But this does not mean that diglossia in Alemannic Switzerland is on its way out; many Swiss, while welcoming the expansion of L-variety domains, see a need to retain domains for Hochdeutsch for a number of reasons. In Singapore, Chinese dialects have lost many domains to Mandarin, but have gained a new one--the religious domain. As many dialect speakers have converted to Christianity, religious services in Hokkien, Hakka, Teochew and Cantonese now serve as the only important public domain for Chinese dialect use.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...domains

This shift was not a slow and `natural' one but was ordered by the government in response to pressures from `democratic' sectors of the society; the church continues to resist, which will probably result in some residual diglossia as long as the church has any influence.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...norm

Some scholars would claim that all languages are diglossic to some extent, so that diglossia would in effect never be eliminated; perhaps at best we can speak of the perception (or to use Fishman's term, the consensus) that diglossia does not exist.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...cultures,

This could also be said for any culture to which Buddhism was exported.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...do.

Another way to define this might be to use the term language reach, a term Pool (1991) uses in discussing the question of the proportion of a population sharing languages. That is, the reach of a language would be the degree of sharedness. One might also refer to the differences between partial and total diglossia as differences in reach.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...franca.

In the complex Tamil linguistic culture already alluded to, there is an educated colloquial style that has become widely disseminated by the medium of the film, and this style is understood and probably actively controlled by all Tamils resident in Tamil Nadu, as well as by Malaysian and Singaporean Tamils, and Indian Tamils in Sri Lanka; it is passively understood by Sri Lanka Tamils and other Tamils domiciled in other parts of India, but not actively controlled by these latter. Sri Lanka Tamils must thus resort to written/formal-spoken to communicate with mainland Tamils, or must switch to English.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.