5.5

5.5 Speech events

One cannot study speech in totality or in abstraction only, one must focus on some clearly definable and delimited segments for analysis if such analysis is ever to be validated, and replicated, if necessary. One such unit of analysis which has been proposed is the speech event. This unit has a beginning and an end, follows a socially recognized patterned sequence, and is generally an entity recognized as such by the people, with a socially accepted designation, for example, a conversation, joke, sermon, interview, prayer, or political speech. Some societies or communities have their own rather unique speech events, and many of these have been studied by the ethnographers of communication. For example, in the United States the activities known as jiving, shucking, playing the frequently studied (see section 9.2).
The investigator studying speech events in a particular community cannot just simply make a detailed list of them and describe each. He must determine the categories which are meaningful to the members of the speech community and the functions they fulfill for them.
Special rules of speaking mark off certain speech events from everyday verbal behavior. These rules involve not only choice of word and topic but also such factor as selection of syntactic and phonological alternates, intonation, speech rhythm, and discourse structure, as well as role and setting constraints. They are often bounded by certain opening and closing routines. For example, if we hear “dearly beloved, we are gathered here today …”we know a wedding is about to be performed, or if we hear “Did you hear the one about the…,”we anticipate that a joke is about to be told. Every child knows that “Once upon a time” announces a fairy tale and that “They lived happily ever after” closes it.
In some types of speech events, co-occurrence restrictions apply much more rigidly than they do in others. For example, public ceremonies or religious rituals prescribe modes of speaking in very narrow terms, whereas in intimate conversation a wide range of alternate sequences is ordinarily permissible. All types of discourse, however, show some form of co-occurrence restrictions. One would be surprised to hear a sentence like I ain’t never gonna analyze no empirical data no more unless someone were trying to be funny. In this case, the humor derives from the fact that we expect “ain’t” to co-occur with informal speech forms and “empirical data” in more formal discourse. Co-occurrence rules in multilingual repertoires tend to be more rigid than in monolingual ones (Gumperz 1971:157).
A major topic concerns the way in which people interpret, that is, make sense out of what is going on in a conversation. There is a particular relationship between what is being said and what is being done, as well as, of course, the social and linguistic context of the conversation. Speakers make assumptions concerning what is going on in the conversation, who says what, what has been said before, and by whom, and whether people are lying, joking, telling the truth, etc. if someone says to us Drop dead!, the first interpretative task is to figure out whether the statement is to be taken literally or not, seriously or not, by evaluating the circumstances. For example, if one were a seriously ill, cardiac patient, it wouldn’t be very funny.
The analyst needs to look at such things as how conversations are begun and ended, as well as the factors determining a person’s right to speak at a specific point in the conversation. He looks at the linguistic means used by speakers to make excuses, convince, cajole, mock, flatter, and so on. There are understandings of how topics may be introduced, avoided, or changed. Lulls and silences have their particular significances.
A conversation begins with an initial utterance by one of the speakers. The letter has a wide choice of expressions to use for this opening utterance which will serve not only to initiate the conversation but also to convey some of his assumptions about this speech and social community and his place in it, as well as the way in which he has conceptualized the social nature of the relationship with the other speaker and the situation in which the conversation takes place. Farb (1974:108-109) claims that all conversations are opened in one of six ways:
1. A request for information, services, or goods
2. A request for a social response
3. An offer of information
4. An emotional expression of anger, pain, joy, which is often a strategy to solicit a comment by a listener.
5. Stereotyped statements, such as greetings, apologies, and thanks
6. A substitute statement to avoid a conversation about a subject the speaker anticipates the listener will broach.
Two utterances are required for either opening or closing a conversation. The second utterance signifies agreement with what the speaker of the first utterance is trying to do, that is, open or close, for example, in American English, “OK,” “I gotta go,” and “well” frequently signal the desire to end a conversation and may be considered “pre-closings.”
It seems patent that all conversation has two basic features, namely, in a given conversation, no more than one person speaks at a time, and in the course of the conversation, the speakers take turns speaking. We may postulate a turn-taking machinery to explain the co-occurrence of these features. this machinery orders speaker turns sequentially in conversation. For example, there are procedures for organizing selection of the next speaker and for determining when and under what conditions transition to a next speaker may or should occur. These procedures operate utterance by utterance rather than being predetermined completely in advance by extra conversational factors. Schegloff (1972b:35) claims that the validity of the rule that one party speaks at a time Is proven by the fact that, where there are four or more persons and more than one person is talking, we can say not that the rule has been violated, but rather that there is more than one conversation going on. But before a conversation is started, there must be a way of determining who is going to speak first. In telephone conversations, for example, the person who answers the telephone speaks first, saying something like “Hello” or “Grubb Construction Co.,” whereupon the caller says something like “This is Joe Gomez. May I speak to M. Ma. Please.” The answerer then says, “This is he,” or “I’ll get him.” Only if someone violates the rules of telephone conversation by discourtesy or by saying something “strange” do we realize that such conversation are patterned, following quite definite rules.
In the midst of rewriting this section of the book, my home telephone rang, I picked up the receiver, and the following exchange ensued (“A”=answerer, “C”=caller):
A: Hello?
C : Hi!
A : Hello!! (somewhat more emphatically)
C : (hangs up)
The caller had obviously reached a wrong number. Once she had ascertained this fact, she terminated the call without enquiry, explanation, or apology. This appears to be not at all unusual in telephone conversations in the United States. There are cultural differences in this regard. Thus, in France, for example, the sequence for the caller at the beginning of a telephone conversation is as follows (Godard 1977).
(1) Check number
(2) Name oneself at the first opportunity
(3) Excuse oneself (optional in case of intimacy)
The French conceive of the telephone as an intrusion, for which apologies are required. In the first place, the caller verifies the number, so that misdialings are caught immediately. It is up to the caller to identify himself first. This is contrary to common practice in the United States. Godard, who is from France, once answered the phone, to the caller’s surprise encountering a nonnative speaker:
A: Hello?
B : Oh! Who are you?
A : Who are you?

It is true that, in the United States, children are often instructed to say something like the following when they call: “This is Johny Jones. May I speak to Jerry, please?” or in answering the phone, to say something like “Cohen residence,” but these norms, if indeed they are norms, appear to be rarely observed. It would appear that more generally speaking, the American caller considers the answerer more a conduit of communication than a person, whereas the reverse is the case in France. Of another call received by Godard in her home in Philadelphia.
A: Hello?
C : Is Jane at your house? Can I speak to her, please?

In France the sequence would be in such a case:
A : Hallo?
C : Checks number
A : Oui
C : Identities himself
A : Greetings. Identifies himself
C : After greetings and a few words, asks for intended audience.

Rather than regarding the call as an intrusion, in the United States people will ordinary interrupt another conversation or any other (or almost any other) activity and answer the telephone. The ring of the telephone has a definite imperative quality. In the United States, a positive value appears to be attached to the act of telephoning it self, and the caller seems to have more rights than the answerer (Godard 1977).
Duncan (1972:283) has described three basic signals for the turn-taking mechanism in conversation, namely (1) turn-yielding signals by the speaker, such as intonation, drawl, body motion, pitch, loudness, completion of a sentence of the use of stereotyped expressions, such as “but uh,” “or something,” or “you know”; (2) attempt-suppressing signals by the speaker, such as movement of the speaker’s hands; (3) back-channel signals by the hearer, such as “mm-hmm” or nods of the head, indicating that he will not take his turn at speaking. Speakers use and respond to these signals in a relatively structured manner.
It often happens that the turn-taking mechanism does not works perfectly, and one speaker interrupts another, or his speech overlaps with another’s. such overlaps and interruptions apparently do not occur in a random fashion but are influenced by interlocutors roles. For example, in one study of eleven cross-sex and twenty same-sex segments of two-party conversations, West and Zimmerman (1974) showed that overlaps were distributed symmetrically in same-sex conversations, but, where the speakers were of opposite sex, males initiated all of the overlaps. There was no difference between the rate at which men interrupted men an women interrupted women, but in the cross-sex conversational segments, 96 percent of the interruptions were initiated by males. The same authors have demonstrated that parent-child interactions are very similar to male-female interchanges in the exercise of interpersonal power. Parents and males interrupt children and females far more than the converse. They conclude, “With respect to conversational interchanges, it is generally the case that the child’s right to speak is problematic and that many of the proprieties and courtesies routinely accorded by adults of equal status are usually ignored in the case of children.” This and similar studies certainly weaken the popular stereotype of women as talking more. Conversational analysis is obviously a source of sociological data on power relationships. For example, instances have been noted particularly in the non-Western world, where a tribal chief or other dignitary will speak through an interpreter to an addressee from another ethnic group, but who understands and speaks the language perfectly well. This practice helps to maintain social distance by preventing direct conversation between the two individuals in question. A similar phenomenon in some modern societies is the custom, still utilized by some women, of not addressing a waiter or waitress in a restaurant directly but, rather, speaking through her escort.
One important theme running through the work of some analysts is the use, by speakers and hearers, of knowledge of the social world in encoding and decoding speech. One interesting and significant problem is formulating place, that is, the means chosen to indicate the physical location by the speaker of some person or object he is mentioning. Some choice must be made, as there are a number of different ways to indicate the “same” place. In a given conversation, however, not all possible formulations would be considered equally correct. Account must be taken of the location of the object, as well as of those taking part in the conversation (Schegloff 1972a). Thus, for example, if someone should ask me where my Xerox copy of Schegloff a article is, in my study, in my files, at home, in California, or back in the States.
In analyzing questions and answers in the context of a conversation, Sacks has identified a rule to the effect that the speaker asking the question has the right to speak if he receives an answer and has aright to ask another question, thus providing for another answer and further question, and so on. Court transcripts are clear examples of such a cycle. Telephone calls are structured in such a way that the person who calls selects the first topic for discussion. In face-to-face social contacts, the speaker uttering the first greeting has the right to talks again when such a greeting is returned and has the right to select the topic (Turner 1970:209-210). In many cases, however, only a simple exchange of greetings may take place.
Shegloff and Sacks (1973:235) reject the notion that their findings relate to some general features of conversational rules in American English, for they do not believe that an ethnic or national language is a relevant putative boundary for their materials and findings. Their approach is to look for standardized invariant rules of discourse, for example, in the sequencing of conversations. To the extent that these rules are culture-specific, however, different rules imply different subcultures and vice versa. The consequences for social interaction in such a case can be significant. As Labov (1971a:64-65) points out, speakers of a given dialect may be able to interpret the grammatical rules of another dialect by may not be able satisfactorily to interpret rules of discourse relating to the interpretation of the social significance of actions, such as in the ways of indication politeness, anger, sincerity, or trust. We interpret sentences rapidly and unreflectingly in terms of social relationships that are not overtly expressed. We are able to do this because of our knowledge of the social system and our familiarity with our interlocutor’s social categories and the cultural associations they carry for him. When we are unaware of these social categories and cultural associations, we may very well misunderstand what is being said.
One of the most extensively studied micro sociolinguistic phenomena is that of code switching, in which a person in the middle of a conversation. Or evening the middle of a sentence, changes from one language, dialect, or style to another. Such switching may take place because the speaker may be able more easily to convey this meaning in the other language, dialect, or style, or the switch itself my convey social meaning.
In order analyze code switching, we must take into account not only what is said, but the social situation in which the conversation takes place, particularly the role relationships and group memberships of the interlocutors, as well as their attitudes toward the various languages, dialects, and styles, and the speakers who habitually use them. Speakers know when to shift from one variety to another. A shift in situation may necessitate a shift in language variety, and conversely, code shifting may indicate a change in the nature of the immediate social relationship of the interlocutors, as well as a change in the nature of the immediate social relationship of the interlocutors, as well as a change of topic or motivation for speech.
These are all cases of situational switching. On the other hand. There is metaphorical switching where there has been no change in situation or topic, but where a person may switch to another language or style to convey humor, warmth, irony, ethnic solidarity, etc.
Scotton and Ury (1977) suggest that a speaker code switches either to redefine the interaction as appropriate to a different social arena or to avoid, by means of repeated code switching, defining the interaction in terms of any specific social arena. Each social arena (e.g. identity arena, power arena, transactional arena) has its own norms concerning the type of behavior expected. It is a strategy by which the skillful speaker uses his knowledge of how language choices are interpreted in his community to structure the interaction so as to maximize outcomes favorable to himself. Thus, in Kenya. For example, a speaker might switch from the local language to Swahili to indicate his perception of the interaction as a transactional, e.g. commercial one, or switch to English as symbolic of power, especially if his interlocutor is weak in English.
Metaphorical switching is possible only because there are previously agreed upon norms governing the use of certain varieties or styles in particular situations. It is in violation of the norm that the switching has its impact. On the other hand, situational switching is called for by the norms-as the situation changes, so does the variety or style which is required. The norms allocate a particular variety or style to particular kinds of topics, places, persons, and purposes. For example, many speakers in the United States shift back and forth between pronunciations like doing and doing’, depending on the status relationship between speakers and hearer, and the topic and setting of the conversation. Such shifting is sometimes beyond the conscious awareness of both speakers and listeners.
In several studies of Chicano code switching, whenever Chicano identity was an underlying theme, Spanish was used by the speaker (for examples of Chicano code switching, see section 9.3). In another study, in a small Norwegian village, Blom and Gumperz (1972) showed that, of the two varieties of Norwegian spoken in the village, the local one was used in issues related to community identification, while the national standard was used in topics more national in scope. It was pointed out, that code switching did not occur in friendly gatherings of people who composed a network of local relationships, regardless of topic, in a situation with both local and non local relationships, however, code switching would be based on topical variation. Speakers conveyed social information by switching from the local dialect to the standard language.