Social Dialect and Language History William Bright Variations usually exist in all levels of linguistic structure in speech communities. These variations can be phonological, grammatical and lexical. Some of them can be connected to geographical locations. For example, there are differences between the English of London and the English of New York. Another type of linguistic variation may depend on the identity of the person spoken to or spoken about. For example, in Nootka there are different linguistic forms used in speaking to or about children, fat people, dwarfed people, hunch backs, etc. Also some variations can be related to the identity of the speaker, like the difference between men’s and women’s speech. In addition to that, linguistic variation is connected to the social status of the speaker. For instance, the difference in speech between the upper-class (U) and middle-class (non-U) in England has now become the only mark of difference between these two classes. There are cases of linguistics variation which are related to the identity of the person spoken to and to the identity of the person speaking. For example, in Yana (an extinct language formerly spoken in north-central California), female speech was used both by women and by men speaking to women. However, male speech was used only by men speaking to men. So, linguistic variation does not reflect the absolute status of either the speaker or the person spoken to. It reflects the relationship between the status of both of them. Also, there are other cases of linguistic variation which are related to factors in the social and cultural context other than the identity of persons. For instance, there are variations that are related to the difference between formal and informal situations. As an example, in American English, pronunciations like huntin’ are found more commonly in informal situations, while pronunciations like hunting are found in formal situations. In other languages, the variations related to formal and informal situations are highly standardized and strictly differentiated. This kind of linguistic variation has been called “diglossia” and it can be found in several languages as Arabic, Swiss German, Haitian French, and modern Greek. It is related to the varying social context within which an individual communicates. A good field for the study of the two types of sociolinguistic variation is the Indian subcontinent. The Indian caste system (the system of dividing society into classes based on differences in family origin, rank, or wealth) helps to recognize the social levels with which linguistic variation is correlated. In the Dharwar District of Mysore state, there are three styles of conversational Kannada. These styles correspond to the three main classes in the social system, which are: the Brahmin, the non-Brahmin, and the Harijan. Moreover, there is a clear difference between formal and informal styles of Kannada. There is a single formal style which all educated people use in certain situations, like lectures, performances and in written composition. The colloquial dialects, the Brahmin and non-Brahmin, have existed in India for a long time, and they are similar enough to be mutually intelligible. Both these dialects of modern Kannada show historical changes from the Old Kannada language. Although they show different changes, the two dialects agree in many changes, such as the loss of medial vowels. There are three hypotheses to explain the identical change in the two dialects: 1- The Brahmin dialect innovated the changes, and the non-Brahmin dialect followed suit. 2- The non-Brahmin dialect innovated the changes, and the Brahmin dialect followed suit. 3- The two dialects developed in the same direction independently. The question is: in the over-all history of a language are changes initiated by the higher social strata or by the lower? To answer the question, there has been a general suggestion that phonetic and linguistic changes are initiated by the upper strata in order to maintain a prestige-marking difference from the lower strata. Then the lower classes imitate their social superiors in order to narrow the gap. And then the upper classes are forced to innovate once more to widen the gap. The Kannada language can be used to test this hypothesis. It was proved that the Brahmin dialect innovates more regarding vocabulary change. For example, the word esar “curry” from Old Kannada became yesru in non-Brahmin dialect, but Brahmin form is huli. In addition, much vocabulary change involved borrowings from Sanskrit or English, and the Brahmin dialect often introduces foreign sounds along with the foreign words. On the other hand, the non-Brahmin dialect shows more sound-change within native vocabulary. Also, on the grammar level, the non-Brahmin dialect seems to have innovated more. In general, the Brahmin dialect seems to show great innovation on the more conscious levels of linguistic change (borrowing and semantic extension), while the non-Brahmin dialect shows greater innovation in the less conscious type of change (phonemic and morphological changes). Therefore, the Brahmin dialect innovates on the more conscious level because of their familiarity with foreign languages. However, it is not apparent why the Brahmin dialect is more conservative than the others in the less conscious type of change. A possible answer is that since literacy is common among Brahmins, the frozen phonology and grammar of the literary language have served to delay change in Brahmin speech. A proof of this hypothesis is in Tulu language which has no writing system. In this language the Brahmin and non-Brahmin dialects show phonemic change in approximately equal degree. But what is the role of social dialects in the process of sound change? It has been suggested that phonemic change is the result of large-scale borrowing from one dialect to another, for example, the contrast between knight and night. It is hypothesized that some members of the English-speaking community may have pronounced knight with an unreleased k. As a result, other speakers, while attempting to imitate the high-prestige dialect, misheard the kn as n and initiated the new pronunciation. This hypothesis can be applied to the Kannada material: when the word pal “milk” in Old Kannada became hal in medieval Kannada, the initial h retained the voicelessness. In modern Kannada, the h of Brahmin halu is partly voiced. It is possible that this change in the Brahmin dialect was misheard by non-Brahmins, and while they are attempting to imitate halu with voiced h, they said alu instead. However, there is no certain evidence that such a process operated in any particular historical change.