Turn-taking is an organization that underlies the regulation and arrangement of the order of speech-making in conversation (Sacks et al. 1974). Under the social norm, age and social status are the standards that govern how people arrange the order.
Albert (1964) had come up with the closest idea before the term “turn-taking” was used. He claims that “the order in which individuals speak in a group is strictly determined by seniority of rank” (p.40-41). A teacher who is younger than the students, for example, has the right to speak first due to the higher rank of teachers in the society. Nevertheless, researchers (Sacks et al, 1974) have found that although lots of changes including different sizes of turns happen during conversation, the transition of the conversation is still “finely coordinated.” In addition, although every conversation has construction composed of various elements like ordered turns and turn sizes in diverse situations, turn-taking is an organized system which “would have the important twin features of being context-free and capable of extraordinary context-sensitivity” (Sacks et al, 1974:699).
Conversation takes place above the sentential level and involves the interaction between two or more individuals with divergent interests and a process of information exchange (Levinson, 1983). In the process of conversation, different speakers take turns to make assessments. People take turns to speak when they are selected. Or they may select themselves to continue speaking while others keep silent (Sacks et al., 1974). As Levinson (1983) suggests, these units are determined by various features of syntactic structure which might be a sentence, clause or noun phrase. Between turns, speakers may find a pause or turning point, which is based on the prediction by the
6
prosodic units or intonation units, to compete for the next turn at the end of the “turn constructional units” which we call transitional relevance place (TRP). Through the observance of TRP, interlocutors can know when to speak and take the turns (Sacks et al. 1974). However, there might be insertion sequences of interruption so that the first and the second part of pairs are not actually adjacent; such as the following question-answer pair, which is interrupted by another pair so that A1 is delayed and not adjacent to Q1:
A: May I have a bottle of Mich? ((Q1)) B: Are you twenty one? ((Q2))
A: No ((A2))
B: No ((A1))
(Merritt, 1976:333)
Natural discourse varies and the context is not as concordant as the written form.
An adjacency pair, therefore, usually contains more than two turns due to unexpected inserted utterances which separate the first and second part of the adjacency pair. But when the second pair is not the expected answer, the preference structure is different.
Question-question, request-refusal, for instance, are in fact not the favorable responses in normal context. Since the response is not preferred and not expected, the speaker who initiates the first part of adjacency pair would try to lead the conversation to his/her expecting direction. However, many other types of response to the question may be selected, apart from the choice of “answer”, such as ignorance, refusals, or challenge to the presupposition of the question (Levinson, 1983:307). Only one acceptable second is the preferred response, but a great number of other alternatives are potential dispreferred seconds. These dispreferred seconds are potential disagreements in the oppositional stance against the prior assessment. There is a
7
ranking between the most preferred and the least preferred seconds, and participants of conversation have many alternative choices among these nonequivalent actions (Lin, 1997). With these nonequivalent alternatives, speakers are able to choose one of them and avoid direct conflict provoked by the disagreements. Levinson (1983: 332) indicates that the idea of preference is not about the individual’s desire, but a structural phenomenon close to the linguistic concept of markedness. In terms of conversants’ expectations, it is observed that preferred seconds are typically unmarked and more acceptable due to the expectation. But dispreferred seconds of different first parts of adjacency pairs share a similarity that they are all marked with various linguistic complexity of preface, delay, accounts and so on.
Disagreement, according to Levinson (1983), is a type of dispreferred seconds and primarily occurs as a response. It happens when a speaker initiates a proposition and the other speaker takes a different stand. The structure of disagreement is complex and mostly preceded with preface, in comparison with the preferred structure.
Indirect disagreement most of the time is mitigated with partial agreement so as to avoid offensive act toward the prior speaker. This type of disagreement is basically equal to so-called “weak agreement/disagreement” according to Pomerantz (1984) since the assessments have been modified in avoidance of aggressive opposition shown through contrastive evaluation. Softening devices like some positive remarks are utilized to save the face of the speaker before expressions of dispreferred disagreement. In the discussion of assessment by Pomerantz, the second assessment is usually relevant with the initial assessment and structured on the basis of the previous assessment to form a pair. Graham (2007:578) has indicated that “all interactions are contextualized and interpreted within the frame of previous interactions and the expectations that grow out of them.” Context is not static but keeps changing in the process of interaction, with the addition of new content of conversation (Sifianou,
8
2012:1561). Therefore, disagreement seems to be a sequential interaction instead of a pair. However, the focal point which is emphasized is the relation between first and its successive assessments.
For the discussion of disagreement, it is necessary to turn back to the discussion of adjacency pairs since disagreement is one type of extensive forms of adjacency pair.
In adjacency pairs like request, invitation, offer, question, hearers have alternatives to either receive or reject them. According to the initial assessment, the second assessment alters with the change of the context. The second assessment may be invited through the first assessment and based on this assessment, the next action chosen by the recipient could be either a “preferred next action” or a “dispreferred next action” (Pomerantz, 1984: 63). Once the second assessment is a bald evaluation with no agreement components or any prefaces, it can be called a strong disagreement, in terms of Pomerantz’s definition. Thus, more varieties are discovered in weak disagreement, such as backdowns, questioning repeats or request for clarification. The delayed statement with prefaced modification also varies with the change of initiative assessment. Two general types are introduced: agreement preferred and agreement dispreferred. Most of the time agreements are preferred undoubtedly. Blimes (1988) explains the agreement preferred situation with participants’ expectations. Agreements are usually invited while speakers performing a speech act. In the circumstance that a speaker performs self-deprecation or self-mockery, however, agreements are dispreferred. An agreement with a prior speaker’s self-deprecation can be interpreted as the criticism to the conversant (Pomerantz 1984:78). Delay devices like silence or hesitation will not be appropriate under such circumstance because silence implies the confirmation of the self-deprecation. In contrast, compliments or disagreements with direct negation are favorable in such a context.
The human interaction contains different opinions and stances, and every single
9
interaction comes with a preference in each interlocutor’s mind. Agreement and disagreement naturally derive from this structure of preference due to the frame of the contextualized interaction. Alternatives between the two poles of preferred and dispreferred response form a continuum of preference structure, from most preferred to least preferred. When the adjacency pair inclines to the dispreferred seconds, the initiative of disagreement develops to a sequence of actions of dispute. Dispreferred responses are always accompanied with potential offensive intention, and this potential aggressiveness manifestly violates the pursuit of harmony in human interaction. The concept of face is being threatened while dispute is provoked. In order to properly manage this “facework”, it is necessary to explain the concept of
“face” and “politeness,” which are introduced in the following section.