Turn-Taking and Disagreement: A Comparison of American English and Mandarin Chinese

This study compares turn-taking and disagreement behaviors in spontaneous conversations in American English and Mandarin Chinese. The English and Chinese speakers observed some turn-taking rules and employed weak disagreement, but differed in the deployment of extended concurrent speech and strong disagreement. Analysis of the Chinese speakers’ reactions reveals nothing negative. This was confirmed by the Chinese speakers’ viewpoints that were explicitly stated in follow-up interviews, which signal that they perceived the practice of extended concurrent speech and strong disagreement in the collected conversations as politic. Furthermore, the similarities and differences between the speakers’ turn-taking and disagreement behaviors appear to be constrained by contextual factors. This discloses the interplay of context, practice, and perception. These findings can raise our awareness of potential issues that might occur in intercultural encounters and the importance of understanding cross-cultural pragmatic differences to avoid miscommunication. compares the turn-taking and disagreement behaviors of the and Chinese speakers who participated in spontaneous conversations in respectively. The non-familial observed some of Sacks, Jefferson’s turn-taking differed in the deployment of extended concurrent of of in


Introduction
The importance of studying the linguistic realizations of speech behaviors in American English versus Mandarin Chinese cannot be overemphasized in the world as it currently exists. Given the present state of trade and diplomacy between the U.S. and China, the potential for misjudgment is ever-present. A lack of knowledge about pragmatic norms in American English and Mandarin Chinese has the potential to lead to misperceptions and misunderstandings in cross-cultural communication between American and Chinese people. It might cause one group to devalue the other group's world views (Corson, 1995), hurt their "self-esteem and group esteem", and "have serious repercussions in situations of world conflict" (Boxer, 2002: 154). Or worse, lack of awareness of pragmatic mismatches may be one reason for discrimination and stereotyping (Gumperz, Jupp and Roberts, 1981;Chick, 1996;Bilbow, 1997) which could severely affect people's lives (Erickson and Schultz, 1982). It can even entail the peril of "alienation" because American and Chinese people "come from distinct linguistic and cultural backgrounds" (Boxer, 2002: 162). Therefore, contrastive pragmatics (Clyne, 1994) is paramount, as Blum-Kulka, House and Kasper (1989) assert. Let us provide a few examples here: 1) It has been said that Chinese speakers are inclined to reject invitations or gifts several times to minimize the cost to others (Gu, 1990) before finally accepting; 2) They tend to say 'suíbiàn 随便.' (It's up to you.) to show thoughtfulness after being asked what they would prefer; 3) They give up their seats to people who are more senior in age or social status to demonstrate respect; 4) In some regions of mainland China, it is common to hear greetings among acquaintances or friends such as 'qùnǎr 去哪 儿?' (Where are you going?) on the street, 'chīlema 吃了吗?' (Did you eat?) around mealtimes, or 'xiàbānle 下班了?' (Have you just got off work?) at the end of the working day when returning home; 5) It might not seem outrageous to receive comments from friends on weight gain, strong suggestions from acquaintances, or questions from strangers about age, marital status, or income. However, these behaviors can appear anomalous to English speakers (e.g., Spencer-Oatey, 2008). As Kádár (2019) states, "the Chinese are often either represented (or represent themselves) as 'super-polite' or as surprisingly rude by Western standards" (p. 203).
In this paper, we examine some similarities and differences between American English and Mandarin Chinese speakers with regard to two speech behaviors: 1) turn-taking, and 2) disagreement. We examine both turn-taking and disagreement at the same time because few prior studies have established a close connection between them. For instance, expressing disagreement can interrupt turn-taking, according to Kern's (2009) research on Japanese speakers; and strong disagreement can cause long overlapping, based on Zhu's (2019) findings about Chinese speakers. Both behaviors are relevant to rules of conversation and (im)politeness. Claims have been made that violations of turntaking mechanisms or disagreement rules are deemed problematic (Schegloff, 2000) or dispreferred (Pomerantz, 1984) in English, although they can be politic in Chinese (e.g., Zhu, 2014aZhu, , 2016Zhu, , 2017a. It has also been assumed that for the purpose of harmony maintenance, Chinese people tend to avoid speech behaviors such as overlapping with seniors (e.g., Lin, 1939) or direct disagreement with higher-status interlocutors (e.g., Ting-Toomey, 1988). Discrepancies in the previous research findings motivate us to reinvestigate these assertions by using empirical data, rather than native-speaker intuitions, relating to both American English and Mandarin Chinese spontaneous conversations among non-familial, relatively equal-status speakers in everyday life.

Turn-Taking
Turn-taking refers to a change of speaker in ordinary conversation. Sacks, Schegloff and Jefferson (1974) present what they deemed to be fourteen facts of turn-taking in any conversation. Of these turn-taking mechanisms, what is most relevant to this study is that people tend to follow the rule of one party at a time to avoid overlapping. Brief overlaps might occur at a transition-relevance place unintentionally and are considered to be "errors and violations" (Sacks, Schegloff and Jefferson, 1974: 701) that should be resolved in fewer than three syllables from the first speaker's utterance completion place (Schegloff, 2000). This minimizes the emergence of long simultaneous speech of more than three syllables from a transition-relevance place. However, in reality, overlapping might not occur at a transition-relevance place. It can last longer than three syllables and hence transgresses Sacks, Schegloff and Jefferson's (1974) orderly turn-taking rules (e.g., Zimmerman and West, 1975). Long overlapping that is initiated by the next speaker more than three words from the current speaker's utterance completion place is called "extended concurrent speech" (Zhu, 2016: 638). In English, three words could be three monosyllabic words or three multisyllabic words. In Chinese, three words are three cí 词 (words) instead of three zì 字 (characters). A Chinese character does not necessarily represent a Chinese word. For instance, biānfú 蝙蝠 (bat) is one cí 词 (word) that has two zì 字 (characters) and dúshū 读书 (to read books) has two cí 词 (words) each of which has one zì 字 (character). Following Schegloff (2000), extended concurrent speech, a rarely explored turn-taking feature, would have been deemed potentially problematic because it is not terminated prematurely before the third syllable of overlapped speech in English.
Overlapped speech in Mandarin Chinese has received less attention than in English. By using Chinese data, some scholars have discussed turn-taking (e.g., H. Liu, 1992Wu, 1997Wu, , 2014Gu, 1999;Liu, 2007;Yang, 2011;Gao, 2012;Li, 2014;Ma, 2014) and interruptions in different discourses (e.g., Ulijn and Li, 1995;Li, 1999Li, , 2001Liao, 2009). H.  and Kuang (2005) tease out the types of interruptions in Chinese. Interruption can lead to overlapping if none of the speakers cease talking. Kuang (2006) presents different types of overlapping, causes of overlapping, and solutions to overlapping, assuming that long overlapping is problematic. In contrast, Zhu (2016) finds that one type of overlapping -extended concurrent speech for floor taking or topic switching -could serve to build, maintain, or enhance interactants' rapport in the everyday talk she collected. The interactants appeared to deem extended concurrent speech for floor taking or topic switching to be an appropriate communicative act, rather than problematic, in post-interaction interviews (Zhu, 2017a(Zhu, , 2017b.

Disagreement
It is natural for people to express opinions in a conversation, and possible responses are either silence, agreement, or disagreement. In this section, we focus on disagreement because, like overlapping that may be interruptive, it is potentially construed as negative speech behavior. Disagreement is reactive, oppositional, and has varying levels of weakness and strength. It is the current speaker's response in interactional communication to the previous speaker's proposition that is verifiable (Takahashi and Beebe, 1993) or untrue (Sornig, 1977;Rees-Miller, 2000). Disagreement does not emerge without an advanced point of view or antecedent verbal (or non-verbal) action. With a proposed point of view or action, though, the current speaker can communicate dissent by delivering differing perspectives (Sifianou, 2012), conflicting stances (Kakava, 1993), disputable positions (Schiffrin, 1985), or contrasting viewpoints (Bardovi-Harlig and Salsbury, 2004). Just like challenges, accusations, threats, and insults, disagreements can be weak and indirect when preceded by pauses, hesitation markers such as uh, repetition of previously-mentioned words, hedges such as well, partial agreements such as it's true that …, concessions such as yeah or you're right, delaying devices, or repair initiators that can soften their tone of voice. A disagreement can also be strong and undeviating if it is "directly contrastive with the prior evaluation. Such disagreements are strong inasmuch as they occur in turns containing exclusively disagreement components, and not in combination with agreement components" (Pomerantz, 1984: 74). That is, strong disagreement and weak disagreement "differ in their relative capacities to co-occur with agreement components" (Pomerantz, 1984: 74). The level of weakness or strength of a disagreement varies on a continuum from weak to strong disagreement.
Many scholars (e.g., Lakoff, 1973;Leech, 1983;Pomerantz, 1984;Brown and Levinson, 1987) have argued that people tend to mitigate disagreement by means of pauses, hesitation markers, repetition of previously-mentioned words, hedges, concessions, partial agreement, delaying devices, or repair initiators in non-self-deprecation cases. Weak disagreement is preferred over strong disagreement because weak disagreement can make people feel more comfortable and satisfied psychologically, whereas strong disagreement has the potential to come across as rude, face-threatening, or offensive (e.g., Beebe and Takahashi, 1989;Kotthoff, 1993). Therefore, politeness can be achieved by maximizing agreement and minimizing disagreement between self and other. There seems to be a widespread tendency for people to attempt to avoid disagreement by expressing regret, "pretending to agree", "displacing disagreement", "telling white lies", or "hedging opinions" (Brown and Levinson, 1987: 117-122).
In terms of research on (im)politeness in Chinese (e.g., Kádár, 2007Kádár, , 2008Kádár, , 2012aKádár and Pan, 2012;Pan and Kádár, 2012;Kádár, Haugh and Chang, 2013), very few studies have focused on strong disagreement in non-familial mundane conversations. Some studies on Chinese interaction have shown optout or weak disagreement to be favored by Chinese speakers for the purposes of maintaining harmony (Bond, 1986;Ting-Toomey, 1988) or avoiding confrontation with people of higher status (Du, 1995;Pan, 2000a;S. Liu, 2004). Ran (2010) explains that direct opposition in Chinese, which contains the structure 不/没有 … (No./There isn't …), or utterance-initial words, such as 不行 (It's not OK), 不对 (This isn't right), 不是 (No, it isn't), and 不许 (No, you can't), can easily cause addressees to have negative feelings and reactions, such as dissatisfaction or anger; it might also result in other antagonistic responses, such as objection and protest; therefore, it is face-threatening and impolite.
Nevertheless, the claim about the universality of disagreement has its pitfalls due to individual, regional, and cross-cultural differences (Clyne, 2006) in the realization of disagreement or the perception of (im)politeness (Chen, 1996;Yu, 1999Yu, , 2003Marco, 2008). Preference for conflict avoidance or weak disagreement varies across cultures (Schiffrin, 1984) and in differing contexts. For instance, after a cultural examination of Jewish people of equal status, both Schiffrin (1984) and Katriel (1986) conclude that disagreement can act as a form of sociability that works to build solidarity, instead of a dispreferred action that threatens social interaction. Likewise, Kakava (2002) discovers that modern Greeks express strong disagreement for solidarity building, while also doing so to create collaborative perspectives (Georgakopoulou, 2001). Zhu (2014a) reveals that strong disagreement in mundane conversations could contribute to the maintenance and enhancement of Chinese face and interpersonal relationships.

A Model of Context, Practice and Perception
We employ Zhu's (2019) model of context, practice, and perception to account for turn-taking and disagreement in the next section, focusing on the sociocultural context, interactional context, and personal context of the study as well as how these contexts constrain the participants' practice and perceptions of turn-taking and disagreement behaviors. The following chart displays the interdependent relationships of context, practice, and perception in general.
T h e s o c i o c u l t u r a l c o n t e x t ( e .g ., s e t t i n g s , r e g i o n s a n d t e m p o r a l i t y ) The interactional context (e.g., goals, risks, topics, relevance, verbal/nonverbal cues, social distance and status differences)

Practice
Perception figure 1 A model of context, practice, and perception zhu, 2019: 14 Figure 1 shows that context, practice, and perception are interdependent on one another. Context comprises the sociocultural context, the personal context, and the interactional context. Specifically, the sociocultural context involves macro-level factors such as the setting, the region, and temporality. The personal context includes individual traits such as sex, age, education, temperament, habits, awareness, and beliefs. The interactional context refers to interactional goals, risks, conversation topics, relevance, interactants' verbal/nonverbal cues, social distance, and status differences. All these contextual factors constrain how people practice and perceive communicative acts. People's habitual practice of communicative acts can help them develop their perceptions of the (in)appropriateness of communicative acts. In turn, these perceptions can help them decide when, where, and how to practice communicative acts.
Consider the sociocultural context of turn-taking or disagreement. How one of these two speech behaviors is realized varies across time and space and in different settings. When and where does it occur? Is it at a formal event? Or is it in an informal conversation? Interactants tend to observe their environment before joining a conversation, let alone conducting extended concurrent speech or expressing strong disagreement. Their different status can result from differences in age, sex, social class, rank, and education, which would cause some adjustment in their turn-taking or disagreement behavior especially in formal settings, although the effect might be blurred in mundane conversations in informal settings. Disagreement employed in mundane conversations in informal settings differs from that which emerges in other formal contexts such as army training (Culpeper, 1996), traffic disputes (Culpeper, Bousfield and Wichmann, 2003), TV shows (Gruber, 2001;Culpeper, 2005), news interviews (Clayman, 2002), mediation sessions (Jacobs, 2002), committee meetings (Kangasharju, 2002), or other contexts (e.g., Pan, 2000a, b;Marra, 2012). Interactants' lack of awareness of the sociocultural context can affect how they deploy verbal/nonverbal cues to take turns or express disagreement properly in social interactions.
The personal context of the interactants makes a huge difference in the practice and perception of turn-taking or disagreement. Are speakers female or male? Are they highly ranked? Are they well-educated? Do they have affable personalities? Do they adopt the convention of talking simultaneously with others or disagreeing directly? Are they aware of the norms of turn-taking and disagreement? What do they believe in terms of (im)polite behavior? If they have formed the habit of using extended concurrent speech or strong disagreement, do they have difficulty dropping it? If they are unaware of alternative norms of turn-taking and disagreement, or if they believe in the appropriateness of their own speech behavior, is it possible for them to change their habits? Interactants' individual traits can shape their practice and perception of communicative acts. This has been well discussed in prior research on speaking turns (e.g., Swacker, 1979) and disagreement (e.g., Walkinshaw, 2007).
Additionally, the interactional context, including goals, risks, topics, relevance, verbal/nonverbal cues, social distance, and status differences, influences the interactants' practice and perception of communicative acts. Interlocutors normally have interactional goals (Spencer-Oatey, 2008) that restrict their speech behavior, such as how to disagree with unequal-status interactants. As might be expected, it has been shown that higher-status interactants tend to disagree with lower-status interactants more directly (Walker, 1987;Fairclough, 1989;O'Donnell, 1990;Locher, 2004), whereas lower-status interactants are more likely to opt out of disagreement or use only weak disagreement (Du, 1995;Pan, 2000a;S. Liu, 2004). But in equal-status situations, if the interactants' goal is simply to socialize, they might enact disagreement without hedges or produce long overlapping, as Zhu (2014aZhu ( , b, 2016 demonstrates. If interactants perform communicative acts in certain routine ways, it can initiate the perception of these acts as normal and appropriate. This is the case because practice and perception might be linked through "the mechanism behind the often observed behavior mimicry and consequent empathic understanding within social interactions" (Chartrand and Bargh, 1999: 905).

Method
In this purely qualitative study, we employ interactional sociolinguistic methods to compare and contrast the two communicative acts that are the focus here -turn-taking and disagreement -in American English and Mandarin Chinese. Interactional sociolinguistic methods are "interpretive methods of discourse analysis", which are used "to gain detailed insights into the many communicative issues that arise in today's social environments, by means of systematic investigation of how speakers and listeners involved in such issues talk" (Gumperz, 2006: 724). This might be one of the earliest approaches to advocate the use of video-taped data in order to incorporate both verbal and non-verbal behavioral features into an investigation and to conduct a micro-level analysis of "the subtleties of miscues" that are "captured in taped interactions" (Boxer, 2002: 152). Watts (2003) also maintains that contextual cues are crucial for interpreting whether certain behavior is appropriate or not. Interactional sociolinguistic analysis looks into both central linguistic features and marginalized contextualization cues, and situates inferences in contexts. Accordingly, its research scope can be extended from minute details of linguistic features to large social/contextual implications. By employing interactional sociolinguistic methods, we examine two datasets -mundane conversations in American English and those in Mandarin Chinese. The first dataset consists of approximately ten hours of audiorecorded conversations by thirty American English speakers. The second dataset is comprised of another ten hours of video-recorded conversations by thirty-four Mandarin Chinese speakers. For the English conversations, two project assistants were hired to collect the data a couple of years ago in a midwestern American city. Provided their English-speaking friends or acquaintances were willing to participate in the project, the researchers asked them to record their everyday conversations in their own social circles. The researchers did not take part in any of the English conversations. For the Mandarin Chinese conversations, the first author was invited, several years ago, by acquaintances or friends to attend informal social events, such as a social meal at a restaurant, a tea party in a condominium, and a social community gathering, in a southeastern city in mainland China. She asked for permission to record the events as research data. She told the participants that the aim of her project was to investigate people's communication style in everyday life and that they should speak in the same way that they normally did. For the majority of the time, she stayed out of the conversations, and was only occasionally drawn into them. Although the recording might have caused the observer's paradox (Labov, 1972), the participants seemed to forget the existence of a recorder as they became increasingly engaged in their conversations. The participants completed a form detailing their demographic information after their conversations had ended. The conversational data were transcribed following Schiffrin's (1987) conventions.
We are not trying to make overgeneralizations when comparing the two datasets. The selected examples in the next section are only representative of the collected data, to which our findings are limited. Since both datasets contained turn-takings initiated by questions, brief overlaps, and weak disagreement, we randomly selected examples of these from the different English and Chinese conversations that were collected. For examples of extended concurrent speech and strong disagreement, we chose examples of the Chinese conversations from between a number of different speakers to ensure as wide a coverage of participants as possible.
We are attempting to uncover the participants' implicit perceptions by analyzing their responses to prior utterances in ongoing conversational interactions. During the conversations, the participants appear to make assessments about each other and respond accordingly, in keeping with Mori's (1999) suggestions. In many cases, assessments of (im)politeness are conducted "on the spot", as interactants draw from certain sets of expectancies in co-constructing interaction in localized, situated contexts (Kádár and Haugh, 2013: 137). If they assess the current speaker's words as appropriate and inviting, they may well enjoy their conversation and actively contribute to it; otherwise, they are likely to show signs of unhappiness or respond negatively (Culpeper, 1996(Culpeper, , 2005Culpeper, Bousfield and Wichmann, 2003;Culpeper et al., 2010;Langlotz and Locher, 2012).
In addition, we interviewed some of the Chinese speakers for their explicit perceptions of extended concurrent speech or strong disagreement. The interviewees we selected were: 1) the speaker whose speech was overlapped or disagreed with, and 2) the speaker who commenced extended concurrent speech or strong disagreement. They were asked whether they were aware of anything abnormal after watching a clip portraying an instance of extended concurrent speech or strong disagreement. They were invited to evaluate the (in)appropriateness of each instance and explain why they thought it appeared that way. Each person was interviewed individually and was promised anonymity to encourage him/her to talk more freely. Each interview lasted approximately thirty minutes and was conducted in Chinese. The data were first transcribed into Chinese and then translated into English. Since we did not locate extended concurrent speech or strong disagreement in the English conversations, we did not interview the English speakers.

6
Results and Discussion

Turn-Taking in English and Chinese
It is safe to say that both the English and the Chinese speakers followed some of Sacks, Schegloff and Jefferson's (1974) turn-taking rules in the collected English and Chinese conversations. For instance, 1) the speakers alternated; 2) turn order, turn size, and the number of parties varied; 3) the length and content of a conversation and the distribution of turns were not predictable; and 4) the speakers also provided prosodic, lexical, and syntactic cues such as rising tones, the pronoun you, or questions to initiate turn-taking.
Excerpts (1) and (2) are examples in English and Chinese, respectively. In Excerpt (1), Ellen, female, and Yasmin, female, are elementary school teachers in their late twenties. They are acquaintances meeting up to socialize. They ask questions to get to know each other better, which naturally leads to turn exchanges.
Excerpt (1)  The pragmatic strategy of asking questions can actively engage interactants in a conversation and give them equal rights to express ideas (Spencer-Oatey, 2008). Given that the context is ordinary social conversation that entails pure interactional talk, we must note that this kind of 'small talk' questioning affects back and forth turn-taking. In a similar way to the English conversations, the Chinese conversations also featured questions as an initiator of turn-taking. In Excerpt (2), Lu, Xiu, Shi, and Jian graduated from the same university fifteen years previously and have been busy building their families and careers. Now they are in their midthirties. Lu, male, is a diplomat working for the Chinese government. Xiu, female, is an English teacher in a private college. Shi, male, is an international business manager. Jian, female, is a Ph.D. student of linguistics. Their first 10.1163/26660393-bja10019 | Contrastive PragmaticS (2021) 1-31 reunion took place in the Chinese city where the data were being collected and their reunion was recorded.
Excerpt (2)  The recording shows that in this restaurant social setting, the interactants try to catch up by asking numerous questions. Their interactional goal of resuming their relationships over a social meal is typical of Chinese culture. As Bian (2001) mentions, Chinese people like to expand their social networks through various activities such as social dining. In excerpt (2), Lu asks questions at turns 1, 3, 5, and 9, and Shi asks questions at turns 4 and 10. These questions can help them fill in their knowledge gaps regarding when Xiu moved house and where she is currently living. It is interesting to see that Shi often overlaps with Lu (turns 4 and 10) when directing his questions to Xiu, but Xiu answers Lu's questions more often. This might be due to the fact that she is sitting closer to Lu and hears him more clearly.
Also, the English speakers often selected the next speaker or let a particular party self-select the next turn in which to speak. In the second case, overlapping sometimes occurred, but it normally ended before the third syllable of the overlapped words.
Excerpt (3) is an instance where brief overlapping appears in the English conversations. In this excerpt, Nancy and Belle, in their early twenties, are two female college students from different departments. They became acquainted after taking some of the same classes. They sometimes lunched together in a residential dining café on campus. Here, they are talking about travel plans for the upcoming summer vacation. After Belle mentioned her plan to visit her relative in San Diego, Nancy became curious about the cost of a flight ticket and initiated a conversation about it.
Excerpts (3)  In this short dialogue, we can see many brief overlaps, including turns 5-6, 7-8, 8-9, 10-11, and 12-13. None of these overlaps go beyond the third syllable of the overlapped words. Also, the interactants often use backchannels that briefly overlap with a couple of the current speaker's words, such as oh, okay, and yeah, to demonstrate good listenership (McCarthy, 2002) and support (see turns 6, 9, 11, and 13). When the next speaker unexpectedly cuts off the current speaker with meaningful concrete words (turn 8), the current speaker stops talking and gives up the floor to avoid long overlapping (turn 7). All these evidently support Sacks, Schegloff and Jefferson's (1974) assertions about turntaking mechanisms. The Chinese conversations also contained many instances of brief overlapping. In Excerpt (4), Mei is a male engineer, in his early twenties, who works for a biotechnology company and plans to pursue a master's degree in Germany. Jia is a female college English teacher in her mid-thirties. They have been acquaintances for many years, although they do not socialize with each other very often due to their busy jobs. As Jia has shared her study-abroad experiences with Mei so that he can be well prepared for his graduate program and new life in Germany, he has invited Jia to a dinner at a restaurant to express his gratitude and bid her farewell.
Except (4)  As we can see from Excerpt (4), Mei and Jia, for the majority of the time, take turns speaking. Occasionally, they briefly overlap. At turn 2, Jia seems to predict the end of Mei's speech a little earlier and starts her turn before he finishes his last words. This is a typical example of what Schegloff (2000) calls "terminal overlaps" (p. 5). At turn 7, Mei acknowledges that he has heard Jia and knows what she wants by using the backchannel oh. This type of overlapping is very common in natural conversations. It indicates the listener's understanding of the speaker and the interactants' compliance with Grice's (1975) cooperative principle.
When it comes to the production of another type of overlapping -extended concurrent speech -the English and Chinese speakers behaved rather differently. While we did not find any instances of extended concurrent speech in the English conversations, in the Chinese conversations we noticed the employment of extended concurrent speech as a pragmatic strategy for active involvement or passionate contribution that can help maintain or enhance interpersonal relationships (Zhu, 2016).
Excerpt (5) is an example of this. Jiao, female, is a full-time Ph.D. student of linguistics in her early thirties. Kang, male, is a part-time Ph.D. student of chemistry in his late twenties. They are strangers who were invited to dinner by a common friend. Prior to this part of the conversation, Kang had mentioned his part-time job with an Education Abroad Consulting company, which initiated a discussion about his payment.
Excerpt (5)  Kang: Customers pay the company. They make a one-off payment … Before Kang completes his explanation about his salary and bonus (turn 2), Jiao cuts in with 客户越多，奖金越高 (The more customers, the higher the bonus) (turn 3) which overlaps extensively with Kang's 是你工资的好几倍 (be several times your salary) (turn 2). Jiao successfully takes the floor away from Kang who, however, does not want to relinquish it until he has finished speaking. The instance of extended concurrent speech is not terminated prematurely or repaired immediately as Sacks, Schegloff and Jefferson (1974) suggest. Kang reacts to it by continuing to explain how he gets paid by the company, the latter being paid by customers before offering any services, which indicates that the extended concurrent speech sounds normal and appropriate to him. The two interactants build the conversation upon each other's contribution, respond to prior utterances without any gapping, and develop ideas vigorously.
After watching the above conversation clip, Kang remarked: I didn't notice anything abnormal in our chat. Yes, we spoke simultaneously, like people always do over informal social meals. But it didn't stop us from communicating. We are both graduate students who were happy to get to know each other by sharing our personal experiences. I was very passionate in telling her about my part-time job. Of course, we talked over the top of each other sometimes. If she cut into the conversation, it meant that she was being attentive. It was not a big deal to be interrupted.
Jiao expressed her opinions in the follow-up interview: Since we were meeting for the first time, I was very curious and asked him lots of questions. When he introduced his part-time job, I was eager to contribute to our conversation by making comments or expressing opinions where they were relevant. If I had waited, I might have forgotten what I initially wanted to say. I wasn't aware that I often interrupted him. But I don't think he cared about that because it was just an informal conversation instead of a formal meeting where you're required to follow an order to take turns. We didn't worry about any consequence of what we talked about and how we talked.

Disagreement in English and Chinese
A close look at the conversational data reveals that the English speakers appeared more agreeable than confrontational in all of the collected conversations. They were inclined to preface disagreement with a pause, concession, hedge, or other type of softeners. Take Excerpt (6) as an example. Larry, a man in his early twenties, has recently graduated with a bachelor's degree. Nicole, a woman in her mid-twenties, had graduated from college several years previously. They are alumni of the same university and have been friends for a few years. Prior to this part of the conversation, Nicole had updated Larry on the progress she was making with job and school applications.
Excerpt (6)  After Larry attempts to associate Nicole's letter of recommendation with her job application to a women's organization (turn 1), Nicole deploys a concession yep and a hedge well before she adamantly says no to express dissent (turn 2). In addition, she follows this with more explanations (turns 2 and 4) to reduce the impact of the dissent, which is well-received by Larry who says Okay (turns 3 and 5). This example supports the arguments of many scholars about the preference which English speakers have for mitigating disagreement with pauses, hesitation markers, repetition of previously-mentioned words, hedges, concessions, partial agreement, delaying devices, or repair initiators (e.g., Leech, 1983;Pomerantz, 1984;Brown and Levinson, 1987). Similarly, weak disagreement occurred in the Chinese conversations, as shown in Excerpt (7) below. Changyu, male, is an automotive engineer. Jiang, female, is a college teacher. They attended the same junior and high schools and are in their mid-thirties. On the day of data collection, they met at their high school classmate reunion and chatted about their lives and other absent classmates. In the following dialogue, Changyu and Jiang discuss one of their former classmates, Nan, who has moved to the U.S.  Jiang: Is it one year? I thought it was three years, maybe. I don't know. I might have mistaken her for another person.
After Jiang mentions that Nan's husband is three years younger than her (turn 2), Changyu tries to deliver an opposing view by gently asking a series of questions 三岁啊？有这么厉害？好像一岁吧？ (Three years? Such a huge difference? Maybe one year?) (turn 3). The rising tone of polar questions, which sound less assertive than outright statements, the utterance-final particles 啊 and 吧, and the lexical choice of the epistemic modality marker 好像 (seem) can all weaken the force of disagreement and make it more appropriate and acceptable to the listener. In response, Jiang starts to doubt her memories and acknowledges that she might have confused Nan with another person (turn 4). This finding endorses Bond's (1986) and Ting-Toomey's (1988) assertions that Chinese speakers strive to maintain interpersonal harmony, which can be achieved by opting out of disagreement or employing weak disagreement (e.g., Du, 1995;S. Liu, 2004;Pan, 2000a). Nonetheless, what cannot be ignored is that the Chinese speakers often deployed strong disagreement to demonstrate their firm standing, although we did not discover any instances of strong disagreement in the English conversations. For example, in Excerpt (8), Xi and Jiu strongly disagree several times during their discussion about networking. Xi is a female English teacher at a private college and Jiu is a female English teacher at a public college. They attended the same college and lived in the same dorm for four years. They have stayed in touch by phone and have met up occasionally since their graduation.
Excerpt (8)  At this meeting, Xi talks about her lack of social adaptability, which, however, meets with Jiu's blunt disagreement (turn 2). The disagreement is enacted in the form of rhetorical questions that do not expect answers, which, in this case, makes the tone of this disagreement stronger than mitigated disagreement. Xi vehemently opposes Jiu's point of view with 不行 (No) at turn 3. While Xi is elaborating on her position regarding her own abilities (turns 3, 5, and 7), Jiu delivers another token of strong disagreement 不要你搞什么关系 (You don't need to network) at turn 8. This then leads to Xi's second strong disagreement 当 然要搞关系 (You definitely should network), which is followed by her long presentation of the importance of networking in China (turn 9). Apparently, Xi and Jiu co-construct strong disagreement and make it the norm in their interaction. Xi responds to Jiu's strong disagreement at turns 2 and 8 with another strong disagreement at turns 3 and 9. After several rounds of strong disagreement, Jiu compromises with 哦 (Oh) at turn 10. The two interactants do not react to strong disagreement with negative evaluations, as they would have done if they had considered it to be impolite (Culpeper, Bousfield and Wichmann, 2003).
In the interview, Xi explained her viewpoint: We've known each other for a while although we don't hang out very often. But when we try to catch up, that is the way we talk. I didn't even notice we hadn't softened our disagreement before you pointed it out. I don't believe it was inappropriate for her to disagree with me directly. She just didn't hide her opinions and I didn't either, which means that we have a good relationship. We got excited about our discussion and we butted heads. So, what? Plus, we were talking about unimportant things, which shouldn't hurt anyone's feelings.
Jiu also shared her perspective as follows: I guess we are two extroverts. Typically, we don't hold back our opinions in informal conversations. We would if we were talking to seniors or to our bosses. But as old friends, we felt comfortable disagreeing with each other directly without causing offence or taking offence. I had confidence in our solid relationship. So, I honestly presented my opinion whenever it was relevant, even though it was contradictory to hers. I might have cut her off sometimes because I couldn't wait. That might be one of my many habits -speaking my mind right away.

6.3
Discussion Pan (2000b) is correct in stating that researchers "should not neglect contextual constraints on the function of each dimension" (p. 145). Contextual factors can affect how closely people follow certain types of norms. Hereinafter, we employ Zhu's (2019) model of context, practice, and perception to account for the aforementioned findings. The broad sociocultural contexts of the collected conversations are contemporary U.S. and mainland China, where linguistic forms and interactional norms can differ from what they were historically. The English conversations took place at apartments (Excerpts 1 and 6) or cafés (Excerpt 3) in a midwestern American city, whereas the Chinese conversations occurred at restaurants (Excerpts 4 and 5), condominiums (Excerpt 8), or community centers (Excerpts 2 and 7) in a southeastern Chinese city. Both are medium-sized cities, which do not have an international reputation like New York or Beijing. All conversations occurred in informal settings, where the participants were not overly concerned about the consequences of the conversations. The ways in which the participants took turns and expressed disagreement were undoubtedly constrained by macro-level factors including temporality, the region, and the setting (Zhu, 2019).
Shaped by similar sociocultural contexts, the English and Chinese conversations share some commonalities. For example, the participants took turns speaking and provided linguistic/non-linguistic cues to start new turns (Excerpts 1 and 2). The number of participants engaging in each conversation altered according to the context. Their conversations differed in length and content, and varied in the order, size and distribution of turns. The participants also briefly overlapped with each other (Excerpts 3 and 4) to manifest active listenership (Knight, 2011). These commonalities apparently reflect some of Sacks, Schegloff and Jefferson's (1974) turn-taking rules. With regard to disagreement, both the English and Chinese speakers employed weak disagreement (Excerpts 6 and 7). The main reason for this finding is that weak disagreement can sound more polite, while strong disagreement might sound more forceful and thus hurt the addressees' feelings.
Furthermore, the interactional context (Zhu, 2019) of the collected conversations, such as interactional goals, risks, topics, relevance, verbal/nonverbal cues, social distance, and status differences, seemed to have influenced the participants' practice of turn-taking and disagreement. For example, all the participants met to socialize with each other, maintain their friendships, or enhance their relationships. They conversed about non-controversial, low-stakes topics that revolved around the trivialities of everyday life, such as traveling (Excerpt 1), residence (Excerpt 2), flights (Excerpt 3), fish (Excerpt 4), salary (Excerpt 5), an application (Excerpt 6), age (Excerpt 7) and social adaptability (Excerpt 8). If the topics had been contentious and high-risk, they might have taken turns or dissented in a different way. The choice of topic affects people's informal reasoning in everyday arguments (Kuhn, 1991). The English speakers were friends (Excerpt 6) or acquaintances (Excerpts 1 and 3), while the Chinese speakers were friends (Excerpts 2, 7, and 8), acquaintances (Excerpt 4), or strangers who joined the conversations through common friends or acquaintances (Excerpt 5). The participants were "on an equal footing to express different points of views" (Angouri andLocher, 2012: 1550). The impact of a status difference that might have resulted from differences in age, sex, social class, rank, or education was undoubtedly mitigated by the nature of mundane conversations in informal settings. The participants' relatively equal status probably played a prime role in the realization of their turn-taking and disagreement behaviors.
To be more specific, the English and Chinese speakers differed in the deployment of extended concurrent speech. The English speakers did not produce extended concurrent speech in their conversations, which indicates that they tried to avoid long overlapping by observing Sacks, Schegloff and Jefferson's (1974) turn-taking rules. But these rules are not necessarily universal. The Chinese speakers produced extended concurrent speech (Excerpt 5). One major reason for this observation could be their high involvement style (Tannen, 1985(Tannen, , 2005(Tannen, , 2007. This high involvement style might have been conditioned by contextual factors such as the informal setting, low-stakes topics, relatively equal status, or short social distance that allowed them to relax and actively engage in the conversations whenever they believed that their contributions were relevant to ongoing topics. The English speakers also differed from their Chinese counterparts in that the latter enacted strong disagreement in addition to weak disagreement. The English speakers did not use any strong disagreement, which supports prior research findings based on English data (e.g., Leech, 1983;Pomerantz, 1984;Brown and Levinson, 1987). In contrast, the Chinese speakers did not refrain from presenting conflicting stances directly (Excerpt 8). As Goodwin, Goodwin and Yaeger-Dror (2002) advance, conflicts "may serve to initiate friendships rather than thwart them" in a different culture or context (p. 1625). Disagreement in Chinese "shows not only the intimate relationship among the speakers but also their strong desire to maintain a sincere and independent self within the interactive frame" (Kuo, 1992: 402). Directness in Chinese could indicate upfront sincerity and solidarity (Lee-Wong, 1994). The ostensibly negative effect of strong disagreement might have been alleviated by the speakers' flat tone, low pitch, and soft volume (Zhu and Boxer, 2013). These observations are aligned with the findings of prior research on the sociability function of disagreement employed by Jews (Schiffrin, 1984;Katriel, 1986) and Greeks (Tannen and Kakava, 1992;Georgakopoulou, 2001;Kakava, 2002).
Finally, the personal context (Zhu, 2019) of the participants must have guided their practice and shaped their perceptions of turn-taking and disagreement behaviors. Culpeper (2005) convincingly argues that research participants are the best evaluators of (im)politeness whose judgments could lead to varying reactions and affect the dynamic of conversational interactions. When the participants in this study completed their conversations courteously after extended concurrent speech or strong disagreement was enacted, it would suggest that they perceived the act as politic, if not polite. If they had perceived an act of strategic impoliteness, they would have responded with negative comments, emotionally loaded words, or signs of distress (Culpeper, Bousfield and Wichmann, 2003). This interpretation was corroborated by the interviewees, who explicitly said that they were not aware of anything abnormal in the Chinese conversation clips that contained extended concurrent speech or strong disagreement. They mentioned the effects of the sociocultural context, such as the informal setting, the interactional context, such as equal status, short social distance, mundane topics, low stakes, and relevance, and the personal context, such as eager personalities, beliefs, awareness, and habits, on their production and views of extended concurrent speech and strong disagreement.
The findings of this study demonstrate a vital link between context, practice, and perception. The various contextual elements seemed to have made extended concurrent speech or strong disagreement in Mandarin Chinese another "type of practice that is open to evaluation as face-threatening, yet appears difficult to characterize as either polite or impolite" Haugh, 2011: 2961). Indeed, which type of practice is "'optimum' depends partly on pragmatic contextual variables and partly on culturally-based sociopragmatic preferences" (Spencer-Oatey andJiang, 2003: 1635). After the Chinese speakers had been exposed to frequent extended concurrent speech or strong disagreement, they did not expect turn-taking to be orderly or disagreement to be mitigated in certain contexts. This understanding would form the basis of the speakers' perceptions and expectations of turn-taking or disagreement that, in turn, could lead to the performance of the two speech behaviors in similar contexts.
The similar and disparate behaviors of the English and Chinese speakers in terms of turn-taking and disagreement imply that the "ways of constructing conversations between two people from different cultures may have similarities and differences" from the "ways in which they converse with members who share their cultural backgrounds" (Austin, 1998: 327). Boxer (2002) rightfully posits that "individuals from two societies or communities carry out their interactions (whether spoken or written) according to their own rules or norms, often resulting in a clash in expectations and, ultimately, misperceptions about the other group" (p. 151). Consequently, this other group would be deemed improper and even rude, although their speech behavior is normal and appropriate in their native society. In other words, communication between people from different cultures runs the danger of being ineffective, and their interpretations of others' intentions incorrect, due to the assumption that people share the same linguistic rules and pragmatic norms. Therefore, although we tend to rely on "implicit understandings and cultural patterns to create and understand messages in conversation" (Austin, 1998: 327), we should develop awareness of different pragmatic norms outside of our comfort zone and acquire knowledge of how to communicate appropriately with people from another society.

Conclusion
This study compares the turn-taking and disagreement behaviors of the English and Chinese speakers who participated in spontaneous conversations in American English and Mandarin Chinese, respectively. The non-familial equalstatus speakers observed some of Sacks, Schegloff and Jefferson's (1974) turntaking rules but differed in the deployment of extended concurrent speech. They employed weak disagreement to lessen the force of disagreement but differed in the deployment of strong disagreement. Analysis of the Chinese speakers' reactions to extended concurrent speech and strong disagreement reveals nothing negative. This was confirmed by the Chinese speakers' viewpoints that were explicitly stated in the follow-up interviews, which signal that they perceived the practice of extended concurrent speech and strong disagreement in the collected spontaneous conversations in informal settings as politic, rather than impolite. In addition to these perceptions, the similarities and differences between the English and Chinese speakers' turn-taking and disagreement behaviors appear to be constrained by macro and micro contextual factors. This discloses the interplay of context, practice, and perception (Zhu, 2019). Also, the findings support Chen, He and Hu's (2013) conclusion regarding the existence of both similarities and differences on different levels in the realization of the requesting behavior of speakers from different cultures. This study can raise our awareness of the potential issues that might occur in intercultural encounters and the importance of understanding cross-cultural pragmatic differences to avoid miscommunication. It is a meaningful addition to studies on contrastive pragmatics. Future research could conduct a quantitative study comparing turn-taking or disagreement in English and Chinese conversations in informal settings to determine whether these similarities or differences are statistically significant. It could also compare turn-taking or disagreement in spontaneous conversations in formal settings to determine whether the formality of settings has an impact on the realization of turn-taking or disagreement by English and Chinese speakers. Investigations along these lines would enrich empirical studies that use authentic data to explore real-life pragmatic behavior. They could also enhance the understanding of intercultural communication between English and Chinese speakers. the editors and the anonymous reviewers for their constructive feedback on earlier versions of this paper. Any remaining errors are our own.
Transcription Conventions (Adapted from Schiffrin, 1987) [ ] Overlapping utterances = Contiguous utterances after an interruption … Omission _ A short untimed pause @ Laughter italics Emphasis (coughs) Characteristics of the talk (indistinct) Items in doubt