Australian Linguistic Society
Overlapping in Australian and Chinese Conversations
Xudong Deng, Department of Language Studies, Edith Cowan University, Perth, firstname.lastname@example.org
When a group of friends are having a conversation together, it is not unusual to find that two or more interlocutors are talking simultaneously with one another. Simultaneous speech or overlaps occur not infrequently in casual conversations between social equals, especially between friends. This can be true of both Australian and Chinese, and many other, if not all, cultures. But it is also widely documented in cross-cultural communication literature, that people from different cultures may follow a different conversation norm and can have different conversational styles, of which the use of overlap in conversations is a part (e.g., Wieland 1991). Interactional sociolinguists have shown repeatedly that when speakers with diverse conversational practices interact with each other, communicative difficulties or even miscommunication are most likely to occur, which can further result in negative cultural evaluations and stereotyping.
This study compares the use of overlap by Australian and Chinese speakers in their respective intracultural conversations. It aims to find out how they differ in terms of the frequency of overlap use, the location in which they place their overlap in reference to a possible completion point of an utterance, and the strategies they use to resolve the state of overlap.
2. Characterisation of overlap
There have been two drastically different characterisations of overlap in previous studies of this conversational phenomenon. The first one is to use it as a defining characteristic of interruption and to characterise it as conflict, involving dominance and power assertion. This characterisation probably originated in the clinical and social psychological studies in the fifties and sixties on interpersonal interactions and their relationships with personality traits and disturbances (e.g., Farina 1960). Later it has been widely used in language and gender studies (e.g., Zimmerman & West 1975). These studies set out to show that men interrupted women in male-female conversations significantly more frequently than women did men, and in these researches, interruption (a subtype of overlap) was seen as micropolitics of men establishing dominance and exerting power over women.
The second characterisation of overlap is associated with Tannen's study of conversational styles (1984, 1994). Following Bennett (1981), Tannen takes the view that overlap and interruption do not belong to the same concept. While overlap is a descriptive term, interruption is basically an interpretive category, which involves making a (moral) judgment on the behaviour of the interlocutor (Tannen 1994: 58). Instead of being viewed as a means of conversational dominance, overlap was seen as an important strategy conversation participants use to show involvement in the conversation. But this involvement strategy can be interpreted differently by different speakers with diverse conversational styles. According to Tannen (1994), the use of similar styles enhanced involvement in the conversation and the use of opposing styles led to misinterpretations.
This study employs Tannen's characterisation of overlap in its analysis of Chinese and Australian conversational styles as its data show that almost all overlaps in the conversations under study indicate the conversationalists' 'fine-grained attention' (Jefferson 1983, 1986) to and active involvement (Tannen 1984) in the conversational work. It is the purpose of the study to reveal whether these two groups of people differ in the use of overlap so that any mismatches in this respect can be discovered and potential areas of misunderstandings can be identified when they interact with each other.
3. Cross-cultural study of overlap
3.1 Cultural differences in the use of overlap
For the past two decades, a considerable number of studies have emerged which examine some culture-specific patterns of overlap in conversations. One of the first studies done in this respect is perhaps Reisman (1974). He observed that people in Antigua, West Indies, did not seem to follow any turn-taking rule in conversations with each other. Interruption could occur anywhere at any time during the course of conversation and the interrupter would not be chastised by co-conversationalists. He used the term "contrapuntal conversations" to describe this phenomenon.
More recently, a group of studies have compared the number of actual overlaps in both intra- and inter-cultural conversations of culturally different groups of people. For example, Wieland (1991), in her study of the turn-taking styles of native speakers of French and of American Advanced learners of French when they talk together with each other in dinner table conversations, found that French speakers used interruptions more frequently than American speakers. A few studies have also examined the way overlaps are used in the conversation and how that differs across cultures. Testa (1988), for example, compares the use of overlap by Italian and British speakers. Despite the commonly-held belief of more interruptive behaviour by Italian speakers, she found no differences in the frequencies of overlap uttered by Italian and British speakers. But instead she noticed that the two groups of speakers use different interruptive pre-starts. English speakers use more indirect pre-starts such as 'well' or tokens of agreement whereas Italian speakers overwhelmingly use direct contrastive markers like 'ma' (i.e., 'but').
3.2 Chinese and Australian conversational styles
Very few studies in the literature can be found which investigate Chinese and Australian conversational styles and none seems to exist which compares the two styles. Australian English speakers may be said to resemble American or British English speakers in their conversational styles, at least to the extent that they follow the one-at-a-time turn-taking rule and that overlaps occur mostly as a systematic error (Sacks, Schegloff & Jefferson 1974). Nevertheless, there is some evidence for distinctively Australian ways of speaking. For example, Guy et al. (1986) noted that Australian English speakers often used a high-rising intonation in statements called Australian Questioning Intonation.
As to the Chinese speakers, existing studies seem to show that they are relatively interruptive in conversation in comparison with a number of other cultural groups. Ulijin & Li (1995) studied the use of overlap/interruption in intercultural multimember party business negotiations between Chinese and Finns and also between Chinese and Dutch. They found that Chinese interrupted more and in a more marked way (i.e. not near a possible completion point) both within their culture and in their interaction with Finns and Dutch, than either Finns or Dutch did intraculturally or interculturally. They suggested that Chinese tend to interrupt as a matter of convention of their language and culture. Graham (1993, cited in Ulijin & Li 1995) compared the use of conversational overlaps (interruptions) by people from ten countries in business negotiations and found that Chinese ranked fourth in terms of the number of interruptions among the ten cultures.
But it is not clear how Chinese would compare with Australians in the use of overlap. Would Chinese use more overlaps in their conversations than Australians or vice versa? Where would they place their overlaps with reference to possible completion points? And how would they resolve the state of overlap once it occurs? This study aims to find out the answer to these questions.
The data used for this study are from 30 dyadic conversations: 15 Australian ones in Australian English and 15 Chinese ones in Mandarin Chinese. Both the Australian and the Chinese conversations consist of 5 pairs of female-female dyads, 5 pairs of male-male dyads and 5 pairs of male-female dyads. Each conversation lasted for approximately 20 minutes, but only a 10-minute segment was used for data. The segment was selected randomly, starting from the second minute onwards. The 60 participants (30 Chinese and 30 Australians) are university undergraduate students. The mean age of Australian students is 21.2 and that of the Chinese is 20.8. The Australian students are all Caucasian Australians who have received their primary or secondary education in Australia, and the Chinese students are from P. R China. The conversation was between friends and was conducted in a classroom setting. The participants were given two conversation topics of general interest, but it was emphasised to them that they should feel free to talk about anything they liked.
4.2 Analytic framework
A great number of systems have been developed in the past three decades to classify overlap use. These systems have used a diversity of criteria in their classification including, for example, speaker switch (e.g., Ferguson 1977), semantic content (e.g. Kennedy & Camden 1983), and pragmatic functions (e.g., Goldberg 1990). These criteria are, however, largely the analysts' own categories and the creation of such categories can go on indefinitely, leaving results of the previous studies mutually incomparable.
In contrast to the above studies, conversation analysts derive their categorisation systems from a detailed analysis of the conversation phenomenon in its sequential context and of the conversation participants' own ways of orienting to the phenomenon at issue. Jefferson (1983: 27) studied the onset of overlap in British and American conversation and observed three distinctive but convergent orientations to talk in progress: Transitional (focussing on a turn's completeness), Recognitional (focussing on a delivery's adequacy), and Progressional (focussing on the talk's flow). Lerner (1991: 454), in his study of the compound turn-constructional unit, found that this kind of unit "provides an opportunity space for speaker transition", and thus can also be a locus of overlap. Jefferson & Schegloff (1975) sketched the orderliness of overlap in conversation by examining its onset, resolution and retrieval.
Based on the above-mentioned conversation analytic work, I shall examine overlap onset and overlap resolution in Chinese and Australian conversations and see what similarities and differences exist in these two respects.
4.2.1 Overlap onset
I classify overlap into two general categories: Transitional overlap and Nontransitional overlap. Transitional overlap is one which occurs at a Transitional Relevance Place (TRP), i.e. the end of a turn-constructional unit where turn change from one speaker to another normally occurs (Sacks, Schegloff & Jefferson 1974; for a linguistic characterisation of TRP, see Clancy et al. 1996). An example of Transitional overlap is given below:
(1) F: I just want that to add to my degree. F: [like that I can teach- ] M: [okay. what do you want to teach] in? F: everything.
Nontransitional overlaps are those which occur at a non-transitional relevance place. For example:
(2) A: I don't othinko kids would see violence A: [as like-, .hh they] see it as violence but B: [yeh. as er-, oas-o ] A: it's not really real to them .
4.2.2 Overlap resolution
Jefferson & Schegloff (1975) specified two main procedures speakers employ to resolve a state of overlap and restore talk to a state of one-party-at-a-time. The first one is that one of the parties drops out and stops talking, as in the following example:
(3) B: I was just going for it. 'cos words were just all B: coming to my mind. [and then often-when] I go back, A: ye:[h. a lot of a-] B: (.) it never really happens to me.
This procedure is called 'dropping' in this study. The other procedure is that the speakers persevere by continuing their talk, engaging in a kind of competition for the turn space, as in (4). This procedure is called 'competing' here.
(4) M: there's always a theory, but it's just a theory in M: the end. like, = =you know, [theories F: =hm.= but is it-, [is it-, M: based [on hypo-, hy-, opothesis u mo] F: [if it was jus'-, if it was jus' one theory, F: you could understand it.] you know, but it's not.
5.1 Overall frequency of overlaps
It would generally be expected that Chinese and Australians would differ in terms of the frequency of overlap use as they are from cultural groups exhibiting 'maximum' socio-cultural differences (Porter & Samovar 1994). Specifically, it might be predicted from the previous studies (Ulijin & Li 1995; Graham 1993 [cited in Ulijin & Li 1995]) that Chinese speakers would use more overlaps than Australians. But this does not seem to be borne out by the result of this study. In effect, the total number of overlaps produced by Australians (n=592) is higher, though marginally, than that by the Chinese speakers (n=539).
5.2 Distribution of overlap types
TABLE 1: Distribution of overlaps in Australian and Chinese dyads Transitional Nontransitional Total Australian 419 (71%) 173 (29%) 592 (100%) Chinese 337 (63%) 202 (37%) 539 (100%)
Table 1 shows the number of Transitional and Nontransitional overlaps together with their percentages of the total number of overlaps produced by Chinese and Australian speakers. Two observations can be made from this result. First, both Chinese and Australian speakers use much more Transitional than Nontransitional overlaps, suggesting that the turn-taking mechanism proposed by Sacks, Schegloff & Jefferson (1974) is to a great extent operative in conversation in the two languages. Second, Australians use somewhat more Transitional overlaps (n=419) than the Chinese speakers (n=337) whereas Chinese use slightly more Nontransitional overlaps (n=202) than the Australian speakers (n=173). This is more markedly shown in the percentages of the two overlap types out of the total number of overlaps. Specifically, Chinese speakers use a higher percentage of Nontransitional overlaps (37%) than Australian speakers (29%) while Australian speakers use a greater proportion of Transitional overlaps (71%) than Chinese speakers (63%).
5.3 Overlap resolution
Similar pictures emerge when we compare how overlaps are resolved in the two languages. Table 2 shows the use of the two types of overlap resolution strategies by Chinese and Australian speakers.
TABLE 2: Overlap resolution in Australian and Chinese dyads Competing Dropping Total Australian 406 (69%) 186 (31%) 592 (100%) Chinese 324 (60%) 215 (40%) 539 (100%)
Again two observations can be made on the basis of this result. First, both Australian and Chinese speakers use more Competing than Dropping to resolve their overlaps. Second, whereas Chinese use more Dropping (n=215) than Australians (n=186), Australians use more Competing (n=406) than Chinese speakers (n=324). This is also more clearly shown in the percentages of the two types of overlap resolution out of the total number of overlaps. That is, Chinese use a higher proportion of Dropping (40%) than Australian speakers (31%) while Australians use a higher percentage of Competing (69%) than Chinese speakers (60%).
6. Discussion and conclusion
This study compares the use of overlap in conversation in Chinese and Australian English. Three aspects of overlap use are compared: the overall frequency of overlap, the relative distribution of overlap types and the use of two overlap resolution procedures. Both similarities and differences have been found as regards these aspects of overlap use. Similarities include: 1) Both Australian and Chinese speakers use the same methods in orienting to the onset of overlap and resort to the same two procedures in resolving the state of overlap; 2) they use similar numbers of overlaps; 3) they both start their overlaps mostly at a possible completion point; and 4) they both tend to compete more with their conversation partners for the turn space than concede it to the other speaker in the resolution of the state of overlap. These similarities may indicate a universal nature of the turn-taking mechanism proposed by Sacks, Schegloff & Jefferson (1974). Conversation participants do seem to follow some basic turn-taking rules common to many, if not all, cultures (see Testa 1988). Given similar interactional and sequential environments, similar interactional resources may be resorted to in the management of moment-by-moment contingencies in the interaction. Previous studies in cross-cultural communication have largely ignored aspects of conversational organisation which may be shared by different cultural groups.
This study has also identified two specific differences in the use of overlap by Australian and Chinese speakers. The first one is that Australians use a higher percentage of Transitional overlaps whereas Chinese use a greater proportion of Nontransitional overlaps. The second one is that when overlap occurs, Chinese speakers drop out more to resolve the state of overlap while Australian speakers compete with each other more to get through the overlap. These differences suggest that different groups may use different strategies for doing the same interactional work, in this case indicating involved participation in the conversation. Chinese speakers achieve their involvement by starting their overlap midway in the other speaker's utterance and dropping out quickly when overlap occurs. Australian speakers, on the other hand, signal their involvement by starting their overlap at a possible completion point but persevere through the overlap with their conversation partners. Minute as the difference might seem, it has the potential to result in cross-cultural miscommunication if their intracultural conversational styles were transferred to the intercultural communication situations.
The findings on the Chinese speakers' great tendency to use overlaps at a non-transitional relevance place matches Ulijin & Li's results that Chinese use a great number of marked interruptions (more or less equivalent to the Nontransitional overlaps in this study) in their interaction with the Finns and the Dutch (Ulijin & Li 1995). It does suggest that interactional strategies used in intracultural communication can get transferred to intercultural communication. This reinforces the need for a critical awareness of the differences in conversational styles of different groups on the part of the intercultural interlocutors.
However, caution must be exercised in interpreting the findings in this study, as its subjects are all university students within the age range of 17 to 26. It is not known whether the results can be generalised to other population groups. Besides, there are other factors apart from the cultural factor which also have an influence on the use of overlap. For example, the study noticed great variations in overlap use in both cultures. These within-culture variations are partly due to the gender factor. The study has found, for instance, that Australian female speakers produce more Nontransitional overlaps than male speakers in their single-gender dyads. In addition, there are also evidences in the study that individual differences can also be a strong factor influencing overlap use. This can be indicated, for example, by the fact that one of the dyads exhibited a much higher frequency of overlap use than any other dyad in the data set. Furthermore, the different characteristics of the two languages (i.e., Chinese and English) may also have an effect on the overlap types used by the two groups of speakers. Thus, for example, the unique topic-comment structure of the Chinese langauge may provide more opportunities for Chinese speakers to produce overlap at a non-transitional relevance place. Moreover, methodologically, it has been found difficult occasionally to assign categories to certain cases. But due to space, these specifics cannot be reported in detail here (but see Deng in progress).
As the use of overlap constitutes only one aspect of a speaker's conversational style, other aspects such as pace and pausing, the use of listener responses, and latching without actual overlapping are also in need of study so that the whole picture of the speaker's conversational style can be revealed. Natural conversational data in intracultural communicative situations could be further compared with the results from this study. Finally, intercultural conversational data could be analysed by using the framework set up in this study to see whether differences in conversational styles in intracultural communication do lead to communication difficulties in intercultural encounters.