Competitive points in Mandarin-speaking multiparty interaction: Speakership and epistemics

This study explores the use of hand points in competitive epistemic and speakership environments in Mandarin-speaking conversation. It examines incoming speakers ’ points at a current speaker. This study employs multimodal conversation-analytic methods to analyse 334 min of triadic interaction. 40 points directed at a current speaker by an incoming speaker are the focus of analysis. This study ﬁ nds that these points indicate an upcoming bid for the ﬂ oor, thereby carrying out self-selection for speakership. In addition, the turn introduced by the incoming speaker's point suggests that they know at least as much as the current speaker about the issue at hand, and can implicate epistemic and speakership competition. As such, these points foreshadow delivery of a weakly aligning turn or a disaligning turn, and may accomplish either af ﬁ liation or disaf ﬁ liation. This study generates new knowledge about the interactional functions of points in multiparty conversation, and suggests potential phenomena for cross-linguistic comparison. © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).


Introduction
A next turn in conversation can be distributed by the current speaker.It can also be seized by an incoming speaker's initiative, i.e., gained through self-selection (Sacks et al., 1974).Negotiation of speakership frequently occurs around a transition relevance place (TRP).When the upcoming turn space is negotiated before a TRP arrives, embodied resources are often prominent signals for next-speaker selection, both when completed by the current speaker and when a prospective next speaker self-selects (Blythe et al., 2018;Iwasaki, 2008Iwasaki, , 2009;;Mondada, 2012Mondada, , 2013;;Mondada et al., 2017).Such transitions are collaboratively achieved by interactants (Goodwin and Heritage, 1990;Iwasaki 2008Iwasaki , 2009;;Mondada, 2012Mondada, , 2013;;Vatanen et al., 2021) and often coordinated via the interactants' use of multiple modalities (Auer, 2018(Auer, , 2021;;Li and Ono, 2019;Luke and He, 2019;Mondada, 2014, p. 140).This study focuses on turn transition in competitive environments, i.e., moments when interactants contest or negotiate their respective positioning.It explores the multimodal resources incoming speakers use when self-selecting and, often, when talking in overlap with the current speaker.It focuses on points directed at the current speaker that bid and may compete for speakership, and examines the epistemic configuration that points index alongside talk and other embodied actions.In the following sections, we will introduce the study by discussing competition for speakership and how epistemics and pointing shape the organization of conversation.

Turn-taking, overlap, and competition
Turn-taking provides a universal infrastructure for talking in interaction, and has a shaping influence on the processing and use of language (Levinson, 2016).Turn-taking is essential for distributing opportunities to participate in interaction, and designing and delivering actions (Schegloff, 2007).One of the normative outcomes of turn-taking in everyday conversation is that one party talks at a time (Sacks et al., 1974;Schegloff, 2000).This outcome is the result of organized practices for turn allocation, providing rights for both the current speaker and others to select who will take the next turn, thereby minimizing instances in which more than one party is selected to speak (Sacks et al., 1974).Despite this, overlapping talk frequently occurs in conversation (Kurti c et al., 2013).Some overlaps are more permissible and legitimate (i.e., they are not treated as a normative violation) while others are treated as competing for the right to hold the floor.For instance, overlap occurring around the end of the ongoing turn is often considered legitimate as the turn-in-progress is approaching a TRP, and a current recipient can take the floor and speak next (Jefferson, 1973(Jefferson, , 1986(Jefferson, , 1993;;Sacks et al., 1974;Schegloff, 2007).In this case, speakership is coherently passed on to the next speaker.As well, there are other forms of vocal behaviour that may overlap the speaker's ongoing turn, but are not considered a floor-taking move (Clift, 2016;Jefferson, 2004;Petitjean and Cangemi, 2016;Raclaw and Ford, 2017).For example, response tokens that overlap around the boundaries of a turn constructional unit (TCU) in a multi-unit turn can also be "minimal and unproblematic" (Jefferson, 1986, p. 154).
Self-selection by a prospective next speaker can be initiated early (Auer, 2005;Jefferson, 2004;Vatanen, 2014Vatanen, , 2018;;Vatanen et al., 2021).Vatanen et al. (2021) investigate how incoming speakers can claim epistemic independence by initiating agreeing responses in early overlap with the current speaker's assertive turn-in-progress.As the current speaker's turn develops, the incoming speaker gains more projective resources (e.g., morpho-syntactic structures in TCUs, would-be completed semantic meaning).These make the TCU trajectory more transparent and enable the incoming speaker to analyse the current speaker's position on the topic being addressed.By choosing-to self-select at an early position, the incoming speaker can adopt an affiliative stance on the matter at hand (i.e., a position that is matched with prior speaker) (see Stivers, 2022; see also Lindstr€ om and Sorjonen, 2012) as well as claiming independent epistemic access to it.
Multimodality has also been studied in relation to overlapping talk.For instance, Mondada and Oloff (2011) investigate gaze, gesture, bodily posture, and verbal and vocal resources in sequential organization of overlapping talk and speakership management.They demonstrate the trajectory of the primary speaker's gestures when she is overlapped by others.To maintain speakership, the primary speaker continues to gesture while their talk is overlapped.The trajectory of the gestures can be consistent with those before the overlap or be continuous but with some perturbation (Schegloff, 2000).The primary speaker can suspend gestures during overlap and continue after it.The continuation of gestures often co-occurs with the primary speaker's regaining of speakership and continuation of the previously halted trajectory of talk.If the primary speaker abandons the trajectory of talk after overlap, the trajectory of gestures often ceases too (Mondada and Oloff, 2011).Along similar lines, Li (2014) investigates body leaning when intervening in the current speaker's turn-in-progress during Mandarin-speaking interactions.In this analysis, Li (2014) explores instances in which speakers launch a question that disaligns with the current turn and activity inprogress, i.e., the question-launching speaker's turn resists or does not support what has been made sequentially relevant by the prior turn (Stivers, 2008(Stivers, , 2022)), while often competing for floor with the current speaker.Li (2014) demonstrates that, while the speakers are launching these questions, they often lean their body and temporarily hold this lean until an answer is given.She argues that this leaning embodies the disalignment and disjunctiveness of the question and is a visual resource to mobilize response from the co-participant.Mondada (2013) explores management of turn-taking that defends speakership in the context of persistent overlaps in multiparty interaction in an institutional context, i.e., political meetings.In these interactions, a would-be speaker signals their orientation to speak next to the chairman, who mediates the turn-taking in the meetings all while another participant is talking.The would-be speaker can make visible their orientation to speaking using verbal and embodied resources, such as gaze, handraising, and muttering.Finally, Luke and He's (2019) study on hand raising and arm-taps in Cantonese conversations explored the incipient speakers' embodied and haptic resources in showing readiness to speak next despite the fact that they have not been selected through prior talk.These non-selected incipient speakers bid for speakership by hand raising bids or arm-taps, making recognizable their orientation toward speaking at the next opportunity.The arm-tap, as a haptic resource, may show a greater urgency in requesting for the floor than raising a hand.Luke and He (2019) demonstrated that while raising a hand, the incipient speaker often protrudes their index finger as well, but this may not necessarily be a point directed at the current speaker.
Together, these studies demonstrate that occurrences of overlap and competing talk in conversation are systematically managed moments, with current and incoming speakers employing a variety of practices to claim and shape the floor.Moreover, embodied resources can play a key role in contextualising participants' conduct, forming a "complex multimodal gestalt" (Mondada, 2014, p. 140), i.e., an emergently integrated whole consisting of multiple modalities that are responding to the contingencies of the activity and the interaction.

Epistemics in interaction
Participants in conversation constantly monitor "who knows what" (Heritage, 2012a, p. 49) and "who knows better" (Stevanovic and Per€ akyl€ a, 2014, p.188), i.e., participants' "epistemic territories" (Heritage, 2012a(Heritage, , 2012b) ) and "epistemic primacy/ authority" (Heritage, 2013;Heritage and Raymond, 2005;Wu, 2018a).Heritage (2012b) puts forward "epistemic status" and "epistemic stance" as two basic issues for organizing participation in interaction.Epistemic status refers to the participant's rights to certain domain of knowledge, relative to the co-participant(s), and aligned with enduring aspects of their identity and/or interactional role.Epistemic stance refers to how knowledgeable the participants display they are relative to one another, and what their epistemic status would imply.There is a large, and growing body of research exploring the relationship between epistemics, action, sequence organization, and grammar/turn design (Enfield et al., 2012;Heritage, 2010Heritage, , 2012b;;Heritage and Raymond, 2021).For instance, participants' epistemic states are inherently connected to how participants recognize each other's actions.In English, determining whether someone has produced an assertion or a question may be entirely reliant on the distribution of knowledge between participants (Heritage, 2013).Moreover, Heritage (2012a) has argued that epistemic asymmetry is a key driving force of sequence organization, with differences in the distributions of knowledge providing a basis for some initiating actions (e.g., questions, topic initiations), and driving sequence expansion.
Question design is often discussed in relation to the epistemic order (Enfield et al., 2012;Heritage, 2010Heritage, , 2012b;;Heritage and Raymond, 2021).Questions typically presuppose strong epistemic asymmetry between speaker and recipient, but the grammatical design of a question (or other actions, for that matter) may convey more knowledge on the part of the speaker.For instance, Kendrick (2010) has studied the epistemic configurations indexed through different grammatical forms of questions and statements in Mandarin.He discusses the TCU-final particle ma, which is the question marker of a turn and is reliably translated into English as a polar question.The particle can solicit information that falls within the recipient's epistemic domain (and outside the speaker's).That is, the particle formulates the turn's propositions as confirmable and thereby make an answering response relevant from the recipient in the next turn (see also, e.g., Robinson, 2020).Actions with strong epistemic dimensions have the potential to be contested and interactionally negotiated (e.g., Heritage and Raymond, 2005).Other previous studies on epistemics in Mandarin have focused on lexical-syntactical items.For example, Wu (2018aWu ( , 2018b) ) illustrates that the turn-prefacing interjection aiyou is used by the speaker in her turn to counteract what is addressed by a previous speaker; specifically, the prior speaker has adopted a position on an issue that she is less knowledgeable about.The speaker who produces an aiyou-prefaced turn, foreshadows a more nuanced understanding of the issue at the next opportunity, while not directly disagreeing.That is, aiyouprefacing clears a path for the speaker to undertake a counter-informing, and exert their epistemic authority.Similarly, Hsieh (2018) explored the use of wenti shi "(the) thing is/(the) problem is" (WTS).He claims that WTS indicates the speaker's epistemic authority on the issue being talked about.When it occurs at the turn-medial position, even when a long-lasting pause follows it, the co-participant waits for more talk from the speaker.This talk is projected by WTS itself, but is driven by the speaker's claim to a more knowledgeable state and more rights to talk.With this practice, the speaker maintains the floor, resisting the coparticipant's potential bid.That is, the maintenance of speakership and epistemic positioning can be intertwined.

Pointing in interaction
Scholars working in a number of traditions have analysed the role of points in making reference to places, locations, persons, and objects (Clark, 2003;Kendon, 2004;Kita, 2003;Streeck, 2009).Blythe et al. (2016Blythe et al. ( , 2018) ) indicate that points are principally deictic, and have a variety of functions in interaction.There is a growing body of conversation-analytic studies that have demonstrated how points can be employed to regulate interactional structures, such as managing incipient speakership (Mondada, 2007), mobilizing co-participation during repairs (Dressel, 2020), indexing a co-participant's knowledge of what is being discussed (Blythe et al., 2016(Blythe et al., , 2018)), and spotlighting what was said by the co-participant as the motivation for the current utterance (Yasui, 2023).Much less is known about these varieties of non-referential pointing, and how they may be used to manage speakership and participation in interaction (Kendon, 2004;Yasui, 2023).Mondada (2007) investigates how an incoming speaker anticipates the approaching TRP and points at an object in a shared space with other co-occurring multimodal conduct to self-select as the next speaker.She demonstrates two types of relationships between the incoming speaker's point and the incipient turn; a turn-initial point and a pre-initial turn point.The turn-initial point is often enacted at the initial phase of the incoming speaker's turn-in-progress, displaying incipient speakership.The pre-initial turn point is conducted before the current speaker's turn reaches its completion, making visible the incoming speaker's self-selection as the next speaker.The pre-initial turn point shows the incoming speaker's anticipation of the upcoming TRP and projects the incoming speaker's orientation towards speaking next.
Other studies have demonstrated that points and demonstrative selection are likely sensitive to sequence organization and epistemic status (Blythe et al., 2016;de Dear et al., 2021;Possemato et al., 2021;Stirling et al., 2022).De Dear et al. ( 2021) investigate points toward locations in conversations in Australian Aboriginal languages and English.The points are not only used to deictically indicate distant locations (Blythe et al., 2016), but also invoke common ground about locations among participants despite referential ambiguity in the concurrent talk.In another recent study, Yasui (2023) illustrates that points in interaction can indicate the origins of an incipiently developing line of interaction as well as mobilize co-participants' recipiency.She analyses moments when the incoming speaker points at the current speaker These points spotlight and topicalize the current action of the indicated co-participant, i.e., the current speaker.This then provides a means for the incoming speaker to initiate a new sequence related to the topicalized action or the prior speaker's behaviour.In particular, Yasui (2023) reveals that these points may be employed as a component of teasing actions, with the point persisting as alongside shifts in gaze to mobilize co-participants' recipiency, e.g., to laugh at what the co-participant just did to trigger the tease.On some occasions, the point is not attended to by the intended addressee.In this case, the pointing participant retracts the point and does not develop their incipient turn any further.
Building on the findings of Mondada (2007) and Yasui (2023), the present study will focus on points that are employed by an incoming speaker as they bid for the floor.Like Yasui (2023), we will explore instances where the incoming speaker's point foreshadows a turn that relates to the current line of talk.But, unlike Yasui (2023), the points in focus accompany a range of responsive actions and effect a range of alignment and affiliation outcomes.Crucially, we will uniquely show that these points are implicated in addressing the relative epistemic positioning of participants on the matter at issue, and function to claim (at least) epistemic parity between the prior and incoming speaker.

Method
The data explored in this study are drawn from 334 min of video recordings collected in Australia and China, as per ethical approvals (reference number: 5289).All participants provided written informed consent.15 people were recruited to participate, and were audio and video recorded in groups of three while engaging in conversation.Groups of participants recruited included friends and University classmates.The participants were instructed to talk with one another as they usually would.No topics were nominated and no other constraints were placed on their interactions.
Each recording lasted between 60 and 80 min.Data were transcribed following conversation-analytic conventions (Hepburn and Bolden, 2013).The transcription of talk is in a three-line structure (Wu, 2014(Wu, , 2018a(Wu, , 2018b)).The first line is Chinese pinyin, the romanization system for Mandarin.The second line is word-to-word translation.In this line, the abbreviations describing word class or syntactical component are also employed (Li and Thompson, 1989; see Appendix A).The third line is free translation.Potentially identifying person and place names mentioned in the recordings were replaced with pseudonyms.The transcription of embodied action is placed under the transcription of talk, as shown in Fig. 1 (Mondada, 2018; see Appendix B), with screenshots depicting embodiment also embedded periodically.
After data transcription, gestures were identified as a potential analytic focus and, following preliminary observations on gestures, hand pointing was selected for data analysis.598 index finger point were found in total, which were then progressively examined for candidate social actions.A subset of these points was employed by an incoming speaker at the current speaker.40 points in this type are the focus of the present study employing multimodal conversation analysis methods.

Analysis
In the following analysis, we will demonstrate in detail the role of points concurrent with an incoming speaker's talk.We argue that these points facilitate turn entry, and contribute to positioning the current recipient as an incoming/incipient speaker.By managing a turn transition in this way, the incoming speakers indicate that they know as much as the current speaker about the specific matter being addressed.The analysis that follows presents five extracts in which the incoming speaker points at the current speaker in a variety of speakership and sequential environments.In Extracts 1 and 2, we will see instances where the incoming speaker takes the floor to deliver a weakly aligning action 1 and adopts an affiliative stance with reference to the prior turn (and speaker).In Extracts 3, 4, and 5, the incoming speakers deliver disaligning turns, and variably adopt an affiliative or disaffiliative stance.

Weakly aligning incipient turns
The participants in Extract 1 are Sophie, Amy, and Lucy.They are friends and university classmates.In the weeks before the recording, Sophie and Lucy had taken part, as volunteers, in an international conference, but Amy was not involved.The 1 Alignment refers to the nature and degree of support that a responsive action provides for the course of action it is addressing (see Schegloff, 2007, p. 58e59).
Following Stivers (2022), we adopt a scalar approach to aligning responses, i.e., aligning responsive practices can be differentially supportive.Therefore, with "weakly aligning incipient turns" in Section 3.1, we are referring to responsive turns that offer weak and minimal endorsement of the design and action of the prior turn while still aligning with it (see also, e.g., Schegloff, 2007, p. 165).By contrast, in the following Section 3.2, we will see incipient turns that overtly contest the design and action of the prior turn and, in doing so, disalign from it.
volunteers were divided into teams.Each team took responsibility for a group of researchers.Sophie and Lucy were in the same team and were responsible for the same group of researchers.Immediately prior to this extract, Sophie has been telling Amy that these researchers were from Bolivia.We will focus on Lucy's point toward Sophie at line 3.
As the extract begins, Sophie gazes at Amy, saying women zhe zu "our team" (line 1) and she points at herself (Fig. 2).The target of the point is consistent with the reference in the talk, i.e., the possessive adjective form of "we" e women "our".She then goes on to say chu le ta fuze de nei ge shi meiguo de "only the researcher she was responsible for is from the US" (lines 1e2), and she stretches her hand out and points toward Lucy (Fig. 3).The target of this point is also consistent with the reference in the turn, i.e., who is referred to by the third-person singular pronoun ta "she" e Lucy.
The point produced by Lucy next has a different function.Lucy produces what will turn-out to be a multi-TCU turn (Robinson et al., 2022), beginning with an aligning response token en "mm" and a concurrent nod (line 3), showing an affiliative stance toward Sophie's prior assertion (Gardner, 2001;Jefferson, 2004;Vatanen et al., 2021).She then says qita dou shi boliweiya de "all the others are from Bolivia" (line 3) with a concurrent point toward Sophie (Fig. 4).With this assertion, Lucy makes explicit what Sophie had implied in her prior turn.As Lucy's turn approaches a TRP, Sophie initiates her subsequent turn with dui "yes" (line 4) and a nod, and then provides a full repeat e qita dou shi boliweiya de "all the others are from Bolivia" (lines 4e5), thereby confirming both Lily's turn and the allusion of her own turn (Schegloff, 1996).Here she stresses dou "all", which had not been stressed by Lucy at line 3.
Lucy points to Sophie at a moment where Lucy may have been expected to take the floor.Sophie and Amy gaze to Lucy toward the end of Sophie's turn at lines 1e2, and she is knowledgeable about the matter at hand.However, Sophie has been principally addressing Amy on this matter, and Sophie's claim about the researchers is not news to Lucy.Lucy's point reflects the sequential pressure created by Sophie's turn, and offers a resource for Lucy to respond to Sophie's assertion, while at the same time changing its emphasis from Sophie's knowledgeable position, i.e., placing into focus where most of the researchers were from.As well, her point reflects the TRP created by her acknowledging response token, allowing her move past it, and self-select to produce another TCU.It is also interesting to note how Sophie responds to Lucy's turn.By offering a full repeat, she claims that Lucy's adjusted focus was alluded to in her original turn (Schegloff, 1996), and thereby seeks to re-exert her epistemic authority over the issue.
An incoming speaker's bid for the floor may be resisted by the current speaker.In this case, the incoming speaker holds the point (sometimes, with a variation of hand shape) and seeks an opportunity to speak next.The participants in the following extract are Mark, Lily, and Tina.Lily moved to a coastal province earlier in her life, on account of her parents' occupation.She then moved back inland to the province of her hometown for university, while her parents remained in the coastal province.The participants are talking about the difference in the price of seafood and aquatic products between the coastal provinces, where Lily lived with her parents, and the inland province, where the university is located.We will focus on Lily's point toward Mark at lines 8 and 10, which seeks to launch a telling related to the topic Mark has been developing.
Mark asserts that seafood and aquatic products in the areas along the rivers are very cheap (lines 1 & 3).Lily responds with dui a "yes" (line 4), in which the turn-final particle a indicates the speaker's previous epistemic access to the matter being addressed (Wu, 2004).The response is followed by Lily's initial attempt to present Shaanxi, where they are located, as an example that demonstrates the price difference (lines 4e5).However, midway through her turn-in-progress, Mark takes the floor using a loud voice, presenting a different coastal province as exemplifying how prices in coastal areas differ from those in inland areas (lines 6e7).When Mark's TCU approaches its end e hen pianyi "very cheap" e Lily produces a multi-TCU turn (Robinson et al., 2022), beginning with a response token dui "yes".This response token aligns the course of action in overlap with the current speaker's turn-in-progress (Gardner, 2001;Jefferson, 2004;Vatanen et al., 2021) and shows an affiliative stance toward what has been addressed by the prior speaker.She then stops gazing downward (Fig. 5) and turns to gaze at Mark (Fig. 6), initiating the following TCU with a prefacing tongue click "tch" and a TCU beginning produced in a loud voice e shang-"last-" (line 8).
Concurrent with her aligning and affiliating first TCU, click, and TCU beginning, Lily employs an index finger point at Mark (Fig. 6).Lily's point toward Mark offers strong evidence of her inclination to speak next, and upgrades her previous attempts at lines 2 and 4e5.However, Mark again resists Lily's multimodal attempt at self-selection and continues speaking.At line 7, he quickly initiates a new TCU in a loud voice e ni huilai zhebian "you return here."As he does this, he redirects his gaze from Lily toward Tina (Fig. 7).Tina has made no comment about this issue, and is therefore possibly less knowledgeable than the others.Mark's gazing toward Tina resists Lily's bid for speakership and frames the turn as being addressed to the less knowledgeable party, which may provide him with a firmer ground to continue competing for speakership with Lily.Mark's TCU at line 7 is grammatically complete (although open to add-ons), but is delivered in the context of a comparison between the prices of seafood in the areas that are along rivers and the inland area; or more specifically, between Guangdong and Shaanxi (lines 1e7).Therefore, Mark may continue or may leave the comparison implicit, and for recipients to infer.
Lily cuts off her turn .tchshang-"Tch Last-" (line 8) as she overlaps with Mark.Lily keeps gazing at Mark (Fig. 7), monitoring the ongoing talk.She holds her point as she adjusts her glasses (Fig. 7); meanwhile, Mark proceeds with his TCU.Lily's persistent gaze and point at Mark receive Mark's shift of his gaze toward Lily at the end of his TCU (Fig. 8).As Mark's turn reaches a TRP, Lily moves her hand from her glasses and reenacts the stroke of her point toward Mark (Fig. 8), reinitiating her turn at line 10 (shang zhou-"last week-").As she says this, Lily's gaze turns downward (Fig. 8).Then, Mark turns to gaze at Tina (Fig. 9) and takes the opportunity to complete his comparison of different prices for seafood e chi dou chi bu qi "can't even afford any as a dish" (line 11).Mark's alternation of gaze between Lily and Tina (Fig. s 9e10) as he says this offers some indication of his orientation toward Lily's speakership bid, but he prioritizes completing his comparison.Lily finally gains the floor, and offers a slightly reshaped TCU beginning from her previous attempt e bu shi shang zhou "not in last week" (line 12).She then commences a telling, reporting that her father posted a video of himself netting shrimp in the coastal province; he then sold the shrimp cheaper than in the food market (lines 12e16).This telling is seemingly what Lily had been attempting to say at lines 8 and 10.As such, we can see a clear topical thread between Mark's talk and her telling, i.e., the incipient turn introduced by the incoming speaker Lily's point stays on the same line of talk as what was addressed by the prior speaker (Yasui, 2023).However, it clearly also relates to the experiences of her own relatives, about whom she has epistemic authority and her co-participants know substantially less (lines 12e16).Therefore, her point was projecting a bid for speakership grounded in this epistemic authority relative to the matter being addressed by the current speaker.

Disaligning incipient turns
As analyzed in Section 3.1, the incoming speaker's authoritative epistemic positioning furnishes speakership and epistemic competition with the prior speaker.In the following extracts, the incipient turn projected by the incoming speaker's point targets specific aspects of the design of prior talk; in doing so, they also effect disalignment with prior talk.In some instances, this occurs despite the current and incipient speakers holding congruent positions on the matter at hand, i.e., they are adopting an affiliative stance.As we will see, prior speakers regularly orient to these dimensions of competition through their talk and embodiment.
The participants in Extract 3 are the same as in Extract 1, and they are talking about their experience at the international conference they were also discussing in Extract 1. Lucy says the American researcher for whom she took responsibility spoke Chinese, so her workload was less than that of the other volunteers.The focus of this extract is Sophie's point at Lucy at line 6.
Prior to the extract, Sophie has been reporting that the International Communication Week, in which the international conference occurred, made her exhausted, and this is what her incomplete TCU at line 1 is referring to.Amy's overlapping turn responds to Lucy's claim that she has been attracted to a boy, also prior to this extract.Lucy's wo yidian dou bu mang "I wasn't busy at all" at lines 3e4 responds to Sophie, and recounts her own, contrastive experience at the international conference.She says the reason that she wasn't busy is that, although the researcher she was responsible for is from the US and his English is very good, he is also zhongguo ren "Chinese" (lines 4e5 & 7).Before Lucy completes the concessive clause, Sophie points at her (Fig. 11) and overlaps with multiple sayings of wo zhidao "I know" (line 6).Lucy persists in overlap with Sophie and, as she does so, points at Sophie with her hand supinated (Fig. 12).Lucy's point occurs before her initiation of the main clause e dan ta you shi zh-zhongguo ren "he is also Ch-Chinese" (lines 5 and 7).Sophie's point makes visible her orientation to take the floor to speak before Lucy's turn arrives to a TRP.Lucy's prompt use of a point after Sophie's exhibits that Sophie's verbal and embodied floor-taking actions challenge the current speaker Lucy for the floor.Overlapping with Sophie's repetition of wo zhidao.wo zhidao "I know.I know" (line 6), Lucy rushes through the initiation of the following TCU in the turn e dan ta "but he" (line 5).The sped-up initiation of the TCU and the concurrent point at Sophie reveal Lucy's initiative to continue her turn and maintain speakership.
Promptly after Lucy's turn arrives at a TRP in line 7, Sophie provides the reason why she knows the researcher Lucy was responsible for is Chinese e yin-yinwei dangshi kan ta de mingzi wo cai ta jiu shi ge huayi "becbecause at that time when I saw his name I guess he is exactly a Chinese descendant" (lines 8 & 10).Lucy responds with a confirmation e dui "yes" (line 9) as Sophie continues speaking.She asserts that the researcher is a huayi "Chinese descendant", which refers to a person of Chinese origin who was born in a foreign country.This provides a different formulation to Lucy, who initially described him as zhongguo ren "Chinese person" in line 7.In response, Lucy offers an agreeing response with multiple components, again confirming with dui "yes" before going on to add ta shi ge huaren "he is a Chinese migrant" (line 12).Lucy stresses ren in huaren (line 12), the component that makes the lexical-semantic meaning different from that of huayi (line 10).The embedded correction (Jefferson, 1987) produced by Lucy implicitly corrects Sophie's previous assertion of huayi, i.e., a person who has a Chinese origin and was born in a foreign country, to be huaren, i.e., a person who was born in China and migrated to a foreign country.Sophie produces an en "mm" (line 14) acknowledging Lucy's turn, passing up the opportunity to take a more substantial turn at talk (Gardner, 2001).In overlap with Sophie's production of en "mm", Lucy repeats ta shi ge huaren "he is a Chinese migrant" (line 15).Then, she says ranhou-e-"andum-" (line 17), followed by the resumption of the course of action she had begun developing at line 4 (Local, 2004;Wang, 2017).
In this extract, Sophie points at the current speaker Lucy (Fig. 11) while she is saying wo zhidao "I know".With this point and her competitive vocal responses, Sophie anticipates the direction of Lucy's turn, which she addresses to Sophie using gaze.Interestingly, Sophie's incipient turn (line 6) co-occurring with the point shows an affiliative stance toward what has been addressed by Lucy but disaligns with the prior speaker's informing/assertion (lines 4e5), i.e., her claim of a just-formed epistemic parity.Instead, Sophie overtly defends her previous epistemic stance with multiple wo zhidao "I know" (line 6).Sophie's point demonstrates her knowledgeable stance, provides a basis for her early responses (Deppermann and Schmidt, 2021), and projects her taking the floor to report on why she knew the researcher was in fact Chinese.So, Sophie's talk and point challenge the current speaker's claim to the floor, and project the upcoming talk grounded in her independent knowledge on the topic.
Another interesting point is that we also see Lucy point at Sophie (Fig. 12) as she rushes through the initiation of the following TCU (line 5).In doing so, she makes an effort to prevent the floor from being taken, and supports her claim to another TCU.Together, both Lucy and Sophie's points demonstrate that speakership is in flux, with both seeking out an opportunity to speak again from their current roles.The competition, initiated by Sophie's point, talk, and other resources (see Fig. 11), continues in the talk that follows, through their differing descriptions of the researcher: Sophie with huayi, which is corrected by Lucy to huaren.Sophie's point, combined with other resources constitutively forms the multimodal gestalt by which she competes with the current speaker in terms of speakership as well as epistemic authority.
As in Extract 3, the competitive point in Extract 4 occurs when the incoming speaker immediately starts to talk in overlap with the current speaker.The indicated participant's response in a later turn displays the K-position she pre-supposed the pointing participant to hold.The participants are Jenny, Suzy, and Mary.They are university classmates.They are talking about the worship of Mazu in Jenny's hometown.Mazu is a goddess that is thought to protect lives of sailors, boatmen, fisherman and tourists, ensuring them safe passage on the sea.In the extract, Jenny is trying to remember the name of an actress who played Mazu in a TV series.Suzy points at Jenny and displays her knowledgeable stance about the issue at line 13.However, she does not continue to hold the floor beyond the turn that co-occurs with her point.12e13).That is, like Suzy, Mary claims independent knowledge about Liu Tao playing Mazu.Suzy nods in the following 0.5second silence and also during her production of a prolonged en "mm" (lines 14e15).With eyes wide open, scratching her head (Fig. 17), Jenny poses a question in line 16 which uses the second-person plural pronoun nimen (nimen zenme zhidao?"how do you know?").The second-person pronoun is stressed and realized with a prosodic contour that conveys surprise, suggesting that her co-participants' knowledge was unexpected.In summary, Suzy's point (Fig. 15; line 11) positions her to take the floor at a moment in which Jenny is poised to deliver information about the person she has targeted with her word search.With this turn, Suzy forestalls Jenny's delivery of this news, claims the floor for herself, and demonstrates a strong epistemic stance on the issue.
In Extract 5, Mark, Lily, and Tina are talking about getting their blood tested by pricking a fingertip.Their university organizes an annual health check in which students undertake this blood test.At the beginning of the extract, Lily complains about her previous experience with this blood test, saying it was very painful.We will focus on Mark's point toward Lily at lines 16 and 19, which accompanies a disaligning and disaffiliating response to Lily's turn at lines 13e15.
Lily complains at line 1 that her previous experience of pricking a fingertip for the blood test was very painful.Prior to this extract, Tina made a joke about the small quantity of blood that can be taken from pricking a fingertip compared to larger quantities of blood drawn from elsewhere.She continues this joke at lines 2e3, saying that this small quantity of blood can only be used for 'distinguishing kinship'.(This refers to a paternity test performed in ancient China, whereby a son's blood should merge into the father's blood, held in a container).Mark responds to Lily's assessment with an old saying that explains why people are sensitive to pain on their fingers e shi zhi lian xin "the nerves of the ten fingertips are linked with the heart" (lines 5e6).Lily agrees and emphasizes that it is "super painful" (line 7).Mark responds at line 9 by undermining Lily's claim, indicating that it is not that painful e hai hao ba "it is okay."From this point, Mark and Lily begin to challenge each other's positions more strongly.
Lily partially repeats Mark's prior turn with rising intonation, and challenges him to do it (line 13).Then, at line 15, she proposes to prick him (wo zha ni yixia "I prick you").In overlap, Mark responds with wo zha guo a "I was pricked".The turn-final particle a shows that the speaker has independent epistemic access to the issue being addressed (Wu, 2004).His assertion of his first-hand experience defends his prior comment at line 9 and undermines the basis for Lily's proposal to prick him, i.e., he does not have an adequate appreciation of the pain associated with this needle.As Mark produces this assertion, he points with his index finger at Lily (Fig. 18).Mark then retracts the gesture, placing his right hand under his chin (Fig. 19), repeating his assertion of first-hand experience, this time in the clear (wo yiqian zha guo a "I was pricked before", line 16).
Lily produces an ingressive hiss while slightly shaking her body, enacting the reaction one might make when pricked with a needle (line 17).As she says this, Mark once again points at Lily (Fig. 20) before producing another TCU at 19.He reiterates his claim about experience getting pricked, with more specific temporal detail e jiu chou xue de shihou "that is, when I had blood drawn" (i.e., when the university health check was performed).After the following 0.2-second silence, Lily again hisses, and Mark again says zha guo "(I was) pricked" (line 21) in a slightly lighter voice while gazing at Lily (Fig. 21).Lily then shifts to talk about another, related matter with Mark e ni zhidao ziji xuexing ma? "do you know your own blood type?" (line 24).
Mark's utterance at line 19 jiu chou xue de shihou "that is, when I had blood drawn" (lines 18e19) is a post-posed temporal adverbial clause (Auer, 2015), which adds on to his prior talk at line 16 with more details.Likewise, zha guo "(I was) pricked" (line 21) can be considered an independent clause but it can also be heard as the main clause of the adverb clause of time, whose subordinate clause is the utterance at lines 18e19.Mark's talk at lines 18e19 and 21 continues adding information about the same issue but Lily, to whom he is addressing his talk, does not provide a verbal response.Nonetheless, Mark's points support his strong, disaligning responses to Lily's turn's by indicating his incipient speakership with a disaffiliative stance, and foreshadowing an authoritative epistemic stance; which, as it turns out, relates to his own experience of being pricked with a needle.It is noteworthy that he points twice, with his second point reflecting the expansion and development of this same stance through his talk at line 18e19.In this case, Mark's authoritative epistemic positioning is used to deliver a contrastive stance toward the previous speaker, and to undermine the position she previously put forward.

Pointing, self-selection, and competition
The extracts presented above demonstrate how an incoming speaker may point at the current speaker, projecting an upcoming floor-taking move, i.e., a bid to self-select.The extracts above recall those described by Mondada (2007) in an institutional context (albeit, an informal one).In the everyday context explored in the present study, we found that these points were used in very distinctive environments of speaker transition (or upcoming speaker transition).When used prior to a TRP, incoming speakers used these points to project turn entry at the next TRP (Extracts 2) (Jefferson, 2004;Vatanen et al., 2021).They also used the points to facilitate immediate turn entry alongside the production of substantial talk in overlap with the current speaker (Extracts 3 & 4).By pointing at the current speaker, and combining it with substantial overlapping talk and other embodied actions, the incoming speaker forms a multimodal gestalt that accomplishes selfselection.
As we have seen, this may spark competition with the current speaker, who may rush through to initiate the next TCU in the turn (line 5 of Extract 3), or avoid establishing mutual gaze with the incoming speaker (Extract 2) who in many cases, is also gazing at the current speaker (cf.Weiss, 2018).With such practices, the current speaker may attempt to prevent the floor from immediately being taken by the self-selecting incoming speaker, and may result in the self-selecting party persisting with or re-enacting their point (e.g., line 10 of Extract 2; Fig. 8).

Competing for epistemic authority
The second key purpose of the points in focus are to invoke epistemic authority on the matter at hand (or a related matter).Although the current speaker's turn-in-progress, by virtue of its first-ness, indexes a more knowledgeable position on the matter at hand (Heritage 2012a(Heritage , 2012b;;Stevanovic and Per€ akyl€ a, 2014), the turn introduced by the incoming speaker's point positions them as similarly knowledgeable about this particular issue.These are rather different epistemic claims to the points reported in previous studies that invoked common ground on locational landmarks (Blythe et al., 2016;de Dear et al., 2021;Possemato et al., 2021;Stirling et al., 2022).We have seen that incoming speakers may use this epistemic authority to variously align and affiliate with prior talk.This study found that these points can foreshadow weak alignment using a multi-TCU turn.Incoming speakers may also overtly disalign, claiming epistemic authority over the matter at issue (Extracts 3e5) and making salient problems with the design of the immediately prior turn (Deppermann and Haugh, 2022;Deppermann and Kaiser, 2022).However, it is important to note that speakers regularly adopted a stance that was affiliative, i.e., their position on the matter matched that of the prior speaker.These variations in alignment and affiliation support the analytic claim that these points have a primarily epistemic function; that is, speakers use their epistemic authority to effect a range of alignment and affiliation outcomes.
This study has additionally shown that epistemic symmetry/parity is a strong basis for motivating participation in interaction; in particular, taking a turn-at-talk.Heritage (2012a) argues that epistemic asymmetry is the driving force of sequences of talk, with shifts from K-to Kþ happening over and over.Like Vatanen (2014Vatanen ( , 2018)), we have shown that competition for speakership can be driven by epistemic parity, in the sense the incoming speaker seeks to demonstrate at least equivalent knowledge on the matter at hand.Perhaps the fact that these acts curtail a turn at talk (and maybe a sequence) actually provides further evidence for Heritage's (2012a) claim.

Conclusion
This study has explored points directed at a current speaker by an incoming speaker when the current speaker's turn is in progress.It generates new knowledge about the functions of pointing in Mandarin-speaking multiparty conversation and extends our understanding of pointing as a constitutive part of multimodal gestalts that can dynamically reconfigure participation and the socioepistemic landscape (Mondada, 2014).The findings of this study provide a potential basis for crosslinguistic comparison, which has become an increasingly frequent application of CA methods, e.g., in Interactional Linguistics (Blythe et al., 2016(Blythe et al., , 2018;;de Dear et al., 2021;Possemato et al., 2021;Stirling et al., 2022).In particular, it raises the question of why pointing should be implicated in this kind of interactional work.This is likely to benefit from a comparative perspective to explore how pointing is recurrently employed between languages and cultures, and the semiotic basis for their relationship with speakership and epistemics (cf.Goodwin, 2018).The present study is limited by its principal focus on the coordination of talk and pointing, and the varied sequence and action environments it examined.Future studies should further explore the possibility of systematic relationships between pointing, talk, and other non-vocal semiotic resources for Mandarin conversations (e.g., gaze, facial expression).Future studies will likely also benefit from examining a narrowing range of sequence/ action environments, and exploring alternative vocal or embodied practices for managing competition for speakership and epistemic authority.
CRediT authorship contribution statement Jessie Chen: Writing e review & editing, Writing e original draft, Formal analysis.Scott Barnes: Writing e review & editing, Supervision.Joe Blythe: Writing e review & editing, Supervision.