3. Nature of data to be collected

Much of the data collected by linguists is intended for linguistic description of otherwise little-known languages. The data can be collected from any good speakers of the language under study provided, of course, they can communicate adequately with the researcher. This latter point has implications for a linguistic sound archive since the eliciting medium might not be standard English (or whatever is the national language). It could be any native language known in common by the researcher and the person to be interviewed or it could be a mixture of a number of these. The problems for a linguistic sound archive can be considerable, especially where a transcription for the recording is not available. Frequently the data collected will contribute to 'whole' language descriptions comprising grammars, dictionaries and text collections. Material may also be collected on particular aspects of a language. Sometimes there is no choice in the matter because a language is so moribund that it is difficult to do more than sketch in the details of the sounds of the language and its vocabulary with a general account of word structures (i.e. phonology, lexicon, morphology). In such cases the collector should strive to gather material from as many surviving speakers as possible, including partial speakers who may provide valuable information in a group interview. Areas which require more detailed study such as sentence structure and meaning and the differing use of language in day to day situations (i.e. syntax, semantics, pragmatics) may be focused on according to the interests of the researcher.

Material should be gathered which places the particular language under study in a wider context. Even if material is only recorded from one language, information should be gathered about neighbouring languages. Attitudes to neighbouring languages can give a first approximation to the nature of the linguistic geography of the area and the relatedness of the languages in the area. A statement like 'We can't talk with those to the west; they speak too rough' may reflect the fact that the form of speech to the west is part of a different language or linguistic grouping. The same statement could yield information about metalinguistic terminology in that language: there may be other linguistic features which are described as 'rough'. Metalinguistic terminology (i.e. the special terms used to describe features of a language in that language) can be a useful tool in gaining linguistic insights. When such terms are known it will be easier to involve the speakers in the investigations (see also section 8).

Any language stands in an intimate relation to its culture and society. The life-style, interests and customs of a speech community will be reflected linguistically. There will usually be differences in speech according to the age of the speaker. In some speech communities female speech is markedly different from male speech. There may be special varieties of speech used only in the presence of certain relations4 or used only by people of a certain seniority or status within the group (the seniority being determined in some cases by the extent of their knowledge of ritual). Attention should also be directed towards gathering data on various linguistic styles. A special style may be used for delivering narrative speeches; in ritual settings; talking to strangers; in songs. 5 Material on how the language is used in practice may also be recorded on tape; although careful observation over a long period will be an essential complement to such recordings.

Too little attention has been focused on this area in the past and it is hoped researchers will direct their energies to this most important component of any language. It is one thing to gain basic communicative competence in a language but far more difficult to convey the 'same' message in a culturally appropriate way to the priest, the prime minister, the mechanic and the mother of one's recently estranged spouse!6

  1. For instance, see chapter 3, Dixon, R.M.W. The Languages of Australia; Cambridge: Cambridge University Press; 1980
  2. The range of material that might be collected is indicated by Homes, D. Language in Culture and Society: a Reader in Linguistics and Anthropology; New York: Harper and Row; 1964
  3. See also Bauman, R. and Sherzer, J. (Eds.) Explorations in the Ethnography of Speaking; Cambridge: Cambridge University Press; 1974