CHAPTER THREE A SURVEY OF DIFFERENCES IN SOCIOLINGUISTIC RULES OF SPEAKING
East is East and West is West, and never the twain shall meet …
The Ballad of East and West (Kipling 1889)
In this chapter Hymes’smodel of the ethnography of speaking (1962, 1967, 1972a) is used as a model for the discussion of differences in sociolinguistic rules of speaking. The purpose of the framework developed by Hymes makes it possible to describe underlying rules of speaking and by implication a description of the communicative competence that enables members of a speech community to use and interpret language successfully. His framework offers a way of discovering the social rules and patterns in operation in speech communities. A description of differences, in turn, enables us to account for misinterpretation of speaker intent.
The major portion of this chapter (from 3.2 to 3.10) is devoted to a discussion of examples from the literature that demonstrate how the components Hymes (1962, 1 967, 1972a) proposed for analysis of the speech event are realized differently in various speech communities. The value of a survey of differences in sociolinguistic rules of speaking is described in 3.11. In 3.12 counter-arguments are given to the criticism levelled at ethnography as a mere strategy for describing differences (instead of theory building or theory testing).
The ethnography of speaking as postulated by Hymes
In this section a motivation for the development of Hymes’sframework is sketched.
Hymes (1967: 13) argued that failure to develop a model and taxonomy of sociolinguistic systems would perpetuate the failure of scientific study to address itself to the unity of language and social life. He was concerned that there had not been a theoretical marriage of linguists and anthropologists and that the scientists studying exotic societies were unable to bring linguistic knowledge to bear on their observations. Anthropologists recorded that different communities have rules which determine communicative behaviour. It was known that communities differ significantly in patterns of code-repertoire, code-switching, and generally in roles assigned to language. Hymes hoped to add « comparative speaking » to the body of comparative studies in other fields, e.g. comparative religion and comparative law, to name but two. Saville-Troike ( 1982:6) reports that the earliest sociolinguistic records are those of White (1880) who reported that the Apache form of greeting differs from those of Americans when he was struck by the fact that the Apache do not kiss each other when greeting, even after a long absence, while Jenness described (as far back as 1929) how girls of the Carrier Indian tribe wore a stone labret to remind them that they had to speak slowly and with deliberation. Sterns (1983) suggests that Malinowski was probably the first to record a particular speech situation when, in 1 923, he described in detail the language used in a fishing expedition. In 1930 Volosinov, a Russian linguist, became aware that different kinds of discourse are used in different situations. He wrote, « village sewing circles, urban carouses, workers’ lunchtime chats, etc. each will have their own type. Each situation, fixed and sustained by social custom, commands a particular kind of organization of audience » (in Macdonnell 1989: 1, trans. 1973:97). These studies were the first attempts at describing what Hymes’ethnography of speaking was to do: to provide a coherent universal structure through which language in all its diversity within a culture and across cultures could be studied.
In his model Hymes facilitates the search for data. He sees the ethnography of speaking as a theory of speech and a system of cultural behaviour concerned with the organization of diversity (Hymes 1971 :51 ). He claims (1967:14)that the absence of a theoretical framework may have led to observations « of great sociolinguistic interest » but, lacking precedent and format for their presentation, » … lie fallow as at best a matter for anecdotes ».
Hymes later modified and elaborated on the 1962 model, but early on ( 1962: 111) he acknowledged the influence of Jakobson’s work especially, as well as his « pervasive debt » to Burke, Pike, Sinclair (1951) and Barker and Wright (1955). Hymes’smodel attempts to elucidate what it is one needs to know about a group’sverbal behaviour in order to participate appropriately and effectively in its activities. This framework consists of two major systems for the analysis of discourse:
- a taxonomy of technical terms, and
- a framework of components for the speech event.
- Compare Chapter 1, Table 1 of this thesis for an analysis of the similarities between his model and that of Jakobson.
As Hymes’ taxonomy of technical terms was discussed in considerable detail in Chapter 1 (see 1 .6.1 ), the components will now be described.
In the theoretical framework he created Hymes uses the acronym SPEAKING to list categories or components of speech which need to be taken into account in a descriptive analysis of a(ny) speech event. He claims that at times as many as sixteen or seventeen components have been distinguished, and that the acronym used for eight components is one that makes the set of components mnemonically convenient. In the acronym the variables that could govern the rules of speaking in speech events are as follows:
- S represents the setting
- P is for the participants in the interaction,
- E is for the ends or goals, i.e. the reason for the interaction,
- A stands for act (which in Hymesian terms means the topic and the manner in which it is discussed),
- K represents key, which has to do with the manner or spirit in which something is said (e.g. playful, sarcastic or serious), I stands for instrumentalities, i.e. the channel used,
- N stands for norms, which can be seen as values and how speech acts are realised differently in different speech communities, and finally the
- .G represents genre.
These variables may affect the way in which language is realised in the speech event, but Hymes points out that not all the components in this framework come into play all the time. He adapted the acronym for French, changing it to PARLANT, thereby illustrating that the categories proposed are not mechanical or « fixed » but are merely offered as a mnemonic device to supply a framework that the researcher can refer to easily when considering all possible factors that might be operative in any particular speech situation. Hymes suggests that in French the heuristic set of components might be presented as P for Participants, A for Actes (form or content), R for Raison or Resultat (ends, outcomes), L for Locale (setting), A for Agents (channels or codes), N for Normes, T for Ton (key) and Types (genre.) In his own words, the pattern Hymes proposes « requires discovering a relevant frame or context, identifying the items which contrast within it, and determining the dimensions of contrast for the items within the set so defined » (Hymes 1962:103). The rules of speaking thus established are rules of communication which form part of a whole network of social rules.
In the rest of this chapter the components proposed by Hymes will be discussed in order to highlight differences that could lead to misinterpretation of speaker intent. Conversational exchange in a variety of languages and societies will be surveyed. Speech communities included in the survey are located in both North and South America, Europe, Africa (including South Africa), Australia, Asia, and the Pacific. The survey will indicate how different discourse strategies are realized by native speakers, and in some cases, by non-native speakers in a range of situations. The accumulation over the years of ethnographic information about verbal practices in human groups illustrates how the meaning of such behaviours may vary from one culture to another. These examples will confirm that the principles of how conversation works are far from universal and underline the need for a bringing to book of conversational rules. Hymes (1967:12) pleads for extended empirical work, saying that what is needed is a widely ranging series of sociolinguistic descriptions because a particular model, let alone an integrating theory, is not convincing unless it has met the test of many diverse situations, or a mass of systematic data.
S for Setting
The letter S in the acronym SPEAKING stands for the two components in which any speech event can take place, i.e. setting and scene. Setting refers to the time and place (concrete physical circumstances) in which a speech event takes place and scene refers to the psychological or cultural setting.
The context in which the speech event occurs determines how participants see their role relationships with one another and the manner in which they think of the goals of the interaction. Schiffrin (1987:3) describes the setting as ranging from « cultural contexts of shared meanings and world views, to social contexts through which definitions of self and situation are constructed, to cognitive contexts of past experience and knowledge » and concludes that understanding how language is used and how it is structured depends on consideration of how it is embedded in all of these contexts.
Some examples of how the context in which speech occurs has an influence on what is said and how it is said follows.
Geographical regions and court settings
Geographical regions can influence the choice of variety. Smith observes that one should not expect the same language, English, to be used in the same ways in London and Los Angeles, Manila and Melbourne, or Tokyo or Toronto (1987:4). Greetings in Germany differ in various geographical regions of the German speaking countries: in Bavaria Gruess Gott is as common as Guten Tag in northern Germany (Armstrong 1986:5). The form of the greeting thus varies according to the region. Keenan (1976:68) demonstrates how setting determines rules of speaking in Malagasy society. She describes a society where new information is a rarity which gives its owner prestige. The Malagasy speaker regularly violates Grice’smaxim « be informative » by refusing to impart what he or she knows. In this (much reported) research, the status that new information gives people in a near-subsistence peasant economy, bestows prestige on them in the eyes of other villagers and creates norms of interaction that violate the « be informative » principle. It is the setting, the plateau of Malagasy, that renders the Gricean principle invalid. On the plains of the Camdeboo in the Karoo, on the other hand, the opposite is true because here rules for behaviour and then value system of the community are substantially different. Here everybody knows everybody else’sbusiness to the extent that the telephonist at the telephone exchange could, until his/her recent replacement by an automatic system, be relied upon to say who was visiting whom and would be have been able to locate anyone who was not at home.
In most societies different speech events are clearly demarcated, but for the Yakan, Philippine Moslems on the island of Basilan, it is difficult for an outsider to distinguish when a group of people talking is in fact in litigation or merely in discussion or negotiation. The reason for this is that the Yakan have no distinctive feature that sets this action apart from others because they have no distinctive settings, such as courtrooms where litigation takes place, and the site of a trial is usually the porch of the house of one of the judges. None of the judges wears robes (Frake 1972: 111 ). For the Yakan the Setting component is therefore irrelevant for the genre identified as court settings in other speech communities.
Differences exist regarding what the accused say in court. Frake ( 1972: 129) compares some of the differences between the Yakan and the Eastern Subanun in participation in litigation. The Subanun devote themselves to trivial disputes in scenes of formal festivities whereas the Yakan try relatively serious cases in scenes of informal discussion. All over the world a court setting, by its very nature, compels the accused, under oath, to reveal all in the proceedings, no matter whether all dignity will be stripped away. On the other hand, until recently in Kangaroo (or « People’s ») Courts in South Africa, the accused did not have much of an opportunity of saying anything, because the customary procedure did not apply at all; the accused was sentenced quickly without a hearing, the verdict often Death by « necklace! »
(In the discussion of the findings in Chapter 5 we will see the important role setting plays in the realization of certain speech acts (specifically requests) on the work-floor. We will see that objections raised by speakers of Afrikaans or English concerning a request for articles of clothing, for instance, are considered taboo at work and purported not to be made by speakers of one of the African languages. In these instances the rules of speaking are defined in terms of location or social context, and not in terms of participants or message form. In other words, the S in Hymes’s framework is the defining (« marked ») component of the speech event.)
Telephonic discourse as setting
Although the telephone is a relatively recent invention, definite rules for its usage exist and telephone behaviour varies from culture to culture. The study of telephone conversations has developed from the work pioneered by Schegloff, as he was particularly interested in conversational openings. Schegloff ( 1972:351) identifies the distribution rule that determines that « the answerer speaks first », and in a corpus of roughly 500 phone conversations found that only one conversation did not follow this pattern. Schegloff (1972:374) comments on how « in the replacement of men by machines a small corner of the social world (telephone interactions) has not escaped ».He says:
It is possible, nowadays (1972!] to hear the phone you are calling picked up and hear a human voice answer, but nevertheless not be talking to a human. However small its measure of consolation, we may note that even machines such as the automatic answering device are constructed on social, and not on mechanical principles. The machine’s magnetic voice will not only answer the caller’sring, but will also inform him when its ears will be available to receive his message, and warns him both to wait for the beep and confine his message to fifteen seconds.
Various cultures have different rules of self-identification when answering a telephone. Rings ( 1 989:463)describes the culturally determined convention according to which German speakers answer the telephone by stating their surname. This convention may seem very abrupt when compared to the American and British response which is usually a « hello ». According to Sifianou (1989:527) the initial response in England seems to be functional whereas in Greece the response seems to be interactional, but Vii-Renko (1989:29) reports that the British and Americans are more chatty and casual on the phone compared to Finns, even when talking business.
Finns answer their home phone calls by identifying themselves while Americans stay anonymous until they know the nature of the call. In Norway the caller identifies her/himself first, while in France it appears that the obligation on the part of the caller is so great for having possibly intruded and imposed, that a routine sequence of questions must be initiated, checking as to the right number having being reached before it is proper to open the topic of the call. Schmidt ( 1975) examined interactions between dyads of Egyptians, Americans and Germans and found that only Germans identified themselves; American callers tended to verify that they had reached the right number, but a dozen or more turns were sometimes necessary for Egyptians who seemed unwilling to be the first to be identified. Clark and French ( 1981) examined the closing section of telephone exchanges and found that there was no straightforward correlation between speech behaviour and social reality in their study of telephone exchanges which end, not with the standard Good-bye, but with a Thank you (in Wolfson 1989:97). Rabinowitz reports that thank you is used frequently to mark partings but that there were no good-byes in 89% of face-to-face encounters in her study ( 1983 in Wolfson 1988:29). Have a nice day is gaining currency in South Africa while in Los Angeles good-bye has apparently been replaced with Missing you already! In television programmes produced in America, it is noticeable that speakers seldom say good-bye when they terminate a telephone conversation, a trend that has recently emerged in the South African soap opera, Egoli.
Differing rules exist for the giving of apologies in telephone calls in the USA and France, causing speakers to follow different sets of rules (Godard 1977). The times when a call may be made differ from country to country and visitors to South Africa are surprised to hear that it is not acceptable to phone a colleague late at night, unlike in Europe or America, for instance.
This section is concluded with the observation that there are culture-specific rules for opening and closing conversations on the telephone, open to misunderstanding by speakers of other speech communities.
P for Participants
Social rules are as diverse as the speech situations that prescribe them, and much depends on the relationship between participants, as we will see in this section in which senders and receivers are discussed. The way in which different participants react depends on factors such as distance (social relationships and how these differ), status (power) and how much imposition exists between the participants, but these are influenced by the setting in which the interaction takes place. The context, wider than the immediate palpable setting, may naturally have tremendous influence on the behaviour of participants: in research conducted into encounters between native speakers of English and native speakers of Zulu at a South African university (Chick 1 987), the Zulu speaking students were found to act in a way that can only be termed powerless, and the broader context of the system of apartheid was found to be responsible for the behaviour of students and the perpetuation of negative racial stereotypes.
In this section power and status differences, forms of address, the notion of face, and conversational openings that are found in different societies will be discussed. (Gender differences have not been discussed.)
By participants we traditionally understand the speaker and hearer. However, various configurations of this combination exist, and Hymes also includes audience and spokesman as further categories. In debates there are multiple audiences as well as multiple speakers. When the addressor is a head of state or some other dignitary whose message is read to reporters by a press agent, the addressor may not even be present (Fasold 1990:44). Furthermore, the hearer or receiver of the message and the addressee need not necessarily be human either. Answering machines are gaining acceptability as addressees although many people initially seem to display some resistance at having to resort to a monologue when leaving a message because a machine cannot follow the normal turn-taking pattern.
1. 1 Introduction
1.2 Purpose of the study and expected outcome
1 .3 The hypothesis
1.4 Language course development
1.5 Background to the problem
2.1 The aim of this chapter
2.2 A short background to the discipline lntercultural Communication
2.3 Information theory
2.4 The Hymesian model
3.2. The ethnography of speaking as postulated by Hymes
3.3 S for Setting
3.4 P for Participants
3.5 E for Ends
3.6 A for Act
3.7 K for Key
3.8 I for Instrumentalities
3.9 N for Norms
3.10 G for Genre
3.11 The value of the survey
3.12 Some criticism, and defence of ethnography
4.2 The problematic nature of data collection
4.3 Ethical considerations
4.4 Different perspectives mean different assumptions: contrast between social psychological and conversation analysis
4.5 Actual means adopted have a critical bearing on results obtained
4.6 Recapitulation of the value of elicitation methods
4.7 The solution: triangulation
5.1 Overview of the chapter
5.2 Sources tapped
5.3 The interview phase
5.4 Objectives of this phase
5.5 Pretoria West
5.7 Second quantitative survey
6.0 Overview of this chapter
GET THE COMPLETE PROJECT
LANGUAGE USE IN INDUSTRY