Für neue Kunden:
Für bereits registrierte Kunden:
35 Seiten, Note: 2,7 (B-)
II.Telephone Conversation From A Conversation Analysis Perspective
1. Conversational Basics
1.1. Conversation Analysis versus Discourse Analysis
1.2.1 Definition of Conversation
1.2.3 Adjacency Pairs
1.2.4 Preference Organization
1.2.5 Problems in Conversations
2. Telephone Conversation
2.2.2 Identification and Recognition
2.2.4 Opening Structure
2.3 Topic talk
2.4.1 Terminal Exchange and Pre-closing
2.4.2 Closing Components and Re-opening
2.4.3 Special Closings
3. Mobile telephone conversation
Although we converse almost every day, we never have exactly the same conversation twice. Nevertheless, certain parts of conversations occur in forms which are very alike. They seem to be constructed according to sets of rules. These rules were examined in the 1970’s for the first time.
The mechanisms which govern our conversations are especially observable in telephone conversations. But since the 1970’s, new technologies have come up and society changed.
The aim of this paper is to examine the mechanisms of telephone conversation and how the systems working in telephone conversations have changed since the establishment of the mobile telephone.
For this, the focus on Conversation Analysis as research methodology is explained, before coming to the basic features of every conversation. Following this, telephone conversations are examined according to their structure of opening, topic-talk and closing. Finally, the changes of this structure for mobile telephone conversations are pointed out.
The basis for the observations on mobile telephone conversation is a survey carried out among 20 Canadian citizens and material provided by the participants of the survey.
Two different main methodologies exist for analysing and examining conversation – conversation in general, or telephone conversation in special – from a linguistic perspective: Conversation Analysis and Discourse Analysis. To understand my decision to focus on the discipline of Conversation Analysis, I will shortly point out the main differences and parallels of these methodologies.
Of course, both disciplines examine conversation. But the methods used for this, the thereby resulting findings and the main understanding of conversation differ immensely.
A common aim of Conversation Analysis and Discourse Analysis is to be able to give “an account of how coherence and sequential organization is produced and understood” (Levinson 1983: 286).
Discourse Analysis uses primitive and basic concepts of linguistics for this. It attempts to extend the rules applying to sentences over the boundaries of sentences. The main method of the discipline is the isolation of sets of units of discourse, followed by a formulation of rules according to these units and finally the division of units into well-formed and ill-formed sequences. The conversations are then analysed according to the rules which have been formulated before. This makes the methods of Discourse Analysis an “immediate categorization of restricted data” (Levinson 1983: 287), which means that not the motivation for the form of the data is searched for, but that some parts of data are isolated from their context which could explain their occurrence and give insight into the real intentions and meanings of utterances. This missing insight is also attributable to the field of Discourse Analysis that is used for the analysis of conversation: the speech act theory. (Levinson 1983: 286)
Speech act theory is in the field of the analysis of conversation mainly concerned with the view that “the level of coherence and order in conversation is to be found [..] at the level of speech acts” (Levinson 1983: 288). For this, a syntax which shall explain what kinds of utterances fit together and which utterances require each other, is added to normal speech act theory. One of the main difficulties of this view is that – as we will see – utterances are able to fulfil more than one speech act at a time. (Levinson 1983: 290)
Conversation Analysis, in contrast, is a more “empirical approach which avoids premature theory construction” (Levinson 1983: 286). Natural occurring conversations build the set of data in which recurring patterns of utterances are searched for. In contrast to Discourse Analysis no rules are formulated according to these patterns, but it is attempted to find out why a special utterance was produced and not an alternative one. From this, sequential organization of the conversation and the way in which utterances are designed to fulfil this organization are researched. In Conversation Analysis, it is not important if an utterance is well-formed or not, but rather why the speaker chose such an utterance and what effect it has on the listener. In this set of preliminaries and orientations, two basic steps are used. (Levinson 1983: 287)
The first step can be described as the isolation of systematic features which can be found in a located conversational organization. Then, the orientation to these systematic features is demonstrated by the conversation’s participants’ orientation to them. (Levinson 1983: 318f) Conversation Analysis understands this orientation of the participants as a justification of their findings, as they conclude that the participants unconsciously interpret the utterances themselves in this way as it is shown by their reactions to certain utterances. (Levinson 1983: 295)
The second step is based on two main questions which shall help to achieve a deeper understanding of the systematics. The first of the questions is “what problems does this organization solve?” (Levinson 1983: 319). The answer to this displays the necessity of the organization. The second question is: “what problems does this organization raise?” (Levinson 1983: 319). By answering this question, the relation to other mechanisms can be demonstrated and the existence of other mechanisms and organizations can be justified.
Thus, the first step in Conversation Analytic research is the identification of organizations in conversation, and the second step is an analysis of the necessity of and relation between such organizations.
The difference between Discourse Analysis and Conversation Analysis should makes it obvious that here, conversation can only be observed from a Conversation Analysis perspective as the examination of the organization of conversation and not of conversation as a kind of sentence, is the aim of this paper. Also, Conversation Analysis certainly has contributed more to the understanding of conversation than Discourse Analysis has, as a conversation has to be seen rather as an interactional product – as claimed by Conversation Analysis – than as a structural product like a sentence – as claimed by Discourse Analysis.
When discussing conversation from a linguistic perspective, one has to know how this term is defined in this context. In everyday language use, conversation often is understood as some kind of “civilized art of talk” or “cultured interchange” (Schegloff 1968: 1075). This is not the understanding Conversation Analysis has of conversation, although this definition - like Conversation Analysis - excludes one kind of talk. This excluded kind is the one of organized talk as is to be found in classrooms, churches and many other institutions. (Levinson 1983: 284)
Conversation Analysis defines conversation in a different way. First of all it is stated to be a “familiar predominant kind of talk in which two or more participants freely alternate in speaking” (Levinson 1983: 284) and the “central or most basic kind of language use” (Levinson 1983: 285). This means, that in a conversation several persons speak in the way we do it ourselves everyday, without a special set of rules according to which the participants are to act, like in a courtroom. The conversation meant here can be a conversation between old friends, a short talk between strangers, a call to a business-partner and so on. We converse in this way almost everyday and more than we do in any other way.
This is not enough for a sufficient definition of the linguistic term ‘conversation’. Sacks (in Coulthard 1985: 69) claims that “a conversation is a string of at least two turns” (Coulthard 1985: 69). This feature is essential for a conversation as will be shown later when discussing for example summons-answer structures. Coulthard (1985: 59) states that one important basic fact of conversations is, that the “roles of speaker and listener change” (Coulthard 1985: 59). This is also observed by Schegloff and Sacks (Sacks & Schegloff 1973: 293) who call this “speaker change” (Sacks & Schegloff 1973: 293). Schegloff and Sacks (Sacks & Schegloff 1973: 293) add another important feature to the basics of conversation which was adopted by all Conversation Analysis researchers: “at least, and no more than, one party speaks at a time in a single conversation” (Sacks & Schegloff 1973: 293). Although exceptions exist when no party or more than one party speaks, it can be observed, that this occurs seldom.
These features are the basis for the most fundamental mechanism in the organization of conversations: the turn-taking system.
According to the main features of conversation, different parties speak, but only one party should speak at a time. This is guaranteed by the mechanism of turn-taking. A single participant is allowed to speak for the duration of his turn. This arises the question as to when a speaker’s turn is over.
A Speaker’s turn is constructed of a syntactic unit, a so called Turn Constructional Unit – or TCU. TCU’s can consist of minimal utterances like mhm as well as of long sentences with several clauses. They are flexible syntactic units which can vary immensely in their length. Certain techniques to lengthen your turn exist. (Levinson 1983: 297)
These techniques include the one of incompletion markers. Incompletion markers occur at the beginning of turns and often have the form of subordinators. The subordinators inform other participants that the turn – and with this the TCU - , will not be finished at the first possible point, but that the turn is lengthened. Another form of incompletion markers are devices which structure the turn as for example I’d like to make two points does. (Coulthard 1985: 64)
Another turn-lengthening technique is utterance incompletion. At the possible end of a TCU a clause connector, like and or but, is used, to add something to the turn. This technique is more problematic than the one of incompletion markers, as other speakers might already have projected the end of the TCU before the utterance incompletion and an overlap in talk might occur. In fact, 28% of overlap occur after the use of this technique. (Coulthard 1985: 64)
The point of completion of a turn is the point in a conversation when the speaker can change. Therefore this point is called Transition Relevance Place or TRP. A set of rules, which will be explained in the following paragraphs, lays how the transition is performed.
The rules allow the selection of a next speaker. A current speaker might in his turn address another participant. At the next TRP the current speaker has to stop and the selected participant gets the next turn. This is also called next-speaker pre-selection. When the current speaker does not pre-select, any other participant might select himself as the next speaker and take over at the next TRP. Then, the speaker who starts his turn first gains the right to speak. If neither the possibility of the rule of next-speaker pre-selection – this is to say that no next speaker was pre-selected – was chosen nor any participant decides to take over by self-selection, the current speaker might continue his turn for the length of another TCU. If this possibility was taken up, all rules apply again at the next TRP. (Levinson 1983: 298)
This gives us the full set of four rules: the rule of next-speaker pre-selection, the one of self-selection the one of current-speaker self-selection and in addition to these the rule defining what happens after a current-speaker self-selection.
Turn-taking is a local mechanism as it only functions between two turns. To make sure that everybody gets his turn at talking, and that not more than one, but at least one person is speaking at a time we act unconsciously according to the set of rules of turn-taking.
The rules only govern the transition from the current to the next speaker. An application of the rules on turns after the next turn is not possible, because such a provision would not meet our definition of conversation. The turn-taking system can also be called a local management system as it “operates on a turn-by-turn basis” (Levinson 1983: 297).
As stated in the previous chapter, a current speaker can select a next speaker by addressing him. Such an address is often done in the form of a first pair part of an adjacency pair.
An adjacency pair is a sequence of utterances which has the length of two utterances, is adjacent, produced by different speakers, typed and ordered as a first part and a second part. Adjacency pairs are sequences like question-answer, greeting-greeting or offer-refusal. (Coulthard 1985: 69)
The parts of an adjacency pair are related. They can be observed to belong to each other and the utterance of a first part makes the completion of the pair by uttering a fitting second part necessary. If no second part occurs it is recognizably missing and waited for. This expectation of a second pair part is called conditional relevance. This conditional relevance can replace an adjacency pair’s characteristic of being adjacent. (Levinson 1983: 306)
Adjacency pairs not necessarily always have to be adjacent. Often, the first part and the second part of an adjacency pair are many utterances apart. Between the two parts so called insertion sequences take place. Insertion sequences are used to sort out preliminaries which are important for the performance of the second pair part. (Levinson 1983: 304)
(1) (Levinson 1983: 304)
A: May I have a bottle of beer?
B: Are you twenty-one?
A’s first pair part which consist in the question if he could have a bottle of beer is not immediately answered to with a second pair part. Rather B poses a first pair part himself by inquiring if A fulfils the requirements for buying beer. The information A gives B in answer to this question is required for B to decide on the answer to the first question and according to it give a second to A’s first part. The sequence thus can be described as:
A: first part of main sequence
B: first part of subsequence
A: second part of subsequence
B: second part of main sequence
The main sequence of question-answer is interrupted by an insertion sequence which consists of a question-answer adjacency pair itself. Insertion sequences can have own insertion sequences, too. (Levinson 1983: 306)
Another main characteristic of adjacency pairs is that they are typed. If this was completely true, there would be no use for insertion sequences, as the utterance of a first part would secure the utterance of a particular second part and further inquiries would be unnecessary. That this is not the case can be illustrated by the example above, when not regarding the main sequence as a question-answer sequence, but as a request-refusal sequence. A requests a beer and B refuses giving it. (Levinson 1983: 306f)
It is common knowledge that requests are not always accepted, but may be refused, too. So, an adjacency pair is typed in a way, that a particular first part requires the utterance of a second part which belongs to a certain set of possible second parts. As will be shown in the next chapter, not all second parts are of equal standing.
Second parts of adjacency pairs can be categorized into preferred and dispreferred second parts. Here, preference and dispreference, are linguistic concepts, although a psychological background certainly exists. (Levinson 1983: 307)
The linguistic preference is closely related to the linguistic markedness. Their relation can be summarized as preferred seconds being unmarked and dispreferred seconds being marked. Thus, preferred seconds “occur as structurally simpler turns” (Levinson, 307) and dispreferreds are structurally more complex. The complexity of dispreferred seconds can consist of several features: delay, preface and account. (Levinson 1983: 307)
Delay as characteristic of a dispreferred second can take place in different forms. While preferred seconds are mostly uttered immediately, dispreferred seconds can show a pause before their delivery. That this form of delay is a feature of dispreference is proved by the fact, that this pause is often interpreted as a preface to a dispreferred second by the recipient.
In such a case the performer of the first part may take the possibility of the pause to initiate repair and with this avoid receiving a dispreferred second. Another form of delaying a second part and giving the other a chance of repair is a delay of a second part over turns. This is done by offering repair by the use of repair initiators. (Levinson 1983: 334)