Coordinated action, communication, and creativity in basketball in superdiversity

ABSTRACT This paper examines the complex social space of basketball training sessions at a sports centre in superdiverse inner-city Leeds, contextualising the site in relation to stigmatising discourses that suggest disorderliness and a lack of social cohesion. The microanalysis of video data from the training sessions counteracts these discourses by showing how social orderliness, cooperation, and creativity unfold in the details of interaction. The significance of its contribution lies in its analysis of communication that bridges across semiotic modes, extending the concept of translanguaging to encompass embodied practice. This practice contributes to constituting a small culture within the basketball club.


Introduction and foundational concepts
This paper reports on a phase of a multi-site team linguistic ethnographic project, TLANG, exploring how communication takes place in urban contexts of superdiversity (Vertovec, 2006(Vertovec, , 2007 in four cities in the United Kingdom. We discuss communication in sports, and specifically in basketball team training sessions held in a multicultural neighbourhood in inner-city Leeds. The TLANG project adopts an overall approach that we describe as a structured visual linguistic ethnography, which stresses not only the importance of reflexivity, context, and systematically documented field experience, but also 'the visual and spatial semiotic dimension of meaning, bringing in attention to physical positioning, the semiotic landscape and the written environment' (Baynham, Bradley, Callaghan, Hanusova, & Simpson, 2015, p. 26). In our analysis of sports practice, we find visual and spatial dimensions of meaning-making to be especially salient.
Our paper makes a contribution to the understanding of translanguaging, communication across languages, modes, and cultures at a time of superdiversity when our cities (and increasingly rural areas) become more complex as people and their communicative resources have become more mobile. In the remainder of this introductory section, we outline our initial positions on superdiversity, translanguaging, and culture, as they relate to the subsequent analysis.

Superdiversity
The movement of large numbers of people from diverse backgrounds from all over the world creates spaces where languages and cultures come into contact in new ways. Indeed, the mass movement of people associated with globalisation, coupled with the mobility of the linguistic and semiotic message in online communication, now indicates cultural and linguistic diversity of a type and scale not previously experienced. For example, the concept of superdiversity, first coined by Stephen Vertovec as a description of the 'diversification of diversity ' (2006, p. 3), aims to capture the sense of mass, rapid, and unpredictable movement of people which characterises the current age. Superdiversity as a sociolinguistic concept is not without its critics, not least for its 'Eurocentric worldview' (Piller, 2015; and note that Vertovec was initially referring to the UK context in recent decades), and for its status in terms of its 'unexamined normative assumptions about language' (Flores & Lewis, 2016, p. 121). Nonetheless, it is retained in this paper because as a descriptive term, it enables us to consider superdiverse practices that we might otherwise not have attended to, especially those beyond the linguistic, and to reconsider established understandings of communicative practice and meaning-making. As Blommaert explains: It is the perspective that enables us not just to analyse the messy contemporary stuff, but also to re-analyze and re-interpret more conventional and older data; now questioning the fundamental assumptions (almost inevitably language-ideological in character) previously used in analysis. (Blommaert, 2015, p. 4) Moreover, it also affords us an acceptance of a new paradigm of uncertainty, of movement, and of mobility, characteristic of contemporary life and communicative practice.

Translanguaging in superdiversity
The complexity of superdiversity has inevitably changed the way we communicate, and the way we understand communication. We regard translanguaging as an appropriate lens through which to understand contemporary communication. Originating in the study and promotion of bilingual education in Wales, the term translanguaging has principally referred to how people draw upon their range of linguistic resources that constitute a repertoire, which they deploy as the exigencies of an interaction demand. Research on the TLANG project has highlighted the salience of multimodal resources and embodied action in the meaning-making process. Hence, we emphasise that communicative practice in contexts of superdiversity involves not just the deployment of multilingual communicative repertoires. Our TLANG approach attends to the ways in which people use an array of available semiotic resourceslinguistic and non-linguisticas they negotiate meaning. Translanguaging is therefore often a creative process, and has the potential to be a transformative process: when communication crosses modes of semiosis in a process of resemiotisation (Iedema, 2003), new meanings are made. Understanding translanguaging as encompassing movement across modes of semiosis relates well to the principles of the form of ethnography known as Nexus Analysis, and to the allied Multimodal Interaction Analysis: these provide the methodological tools for the analysis of the data on which we report in this paper.

Cultures big and small
Insofar as translanguaging relates to culture, this paper expounds the notion that the trans-semiotic communication integral to the practice of basketball as it takes place in the club we examine here contributes to the dynamic and ongoing constitution of a local, small culture (Holliday, 1999). This is a culture that emerges as concomitant with transformative translanguaging practices and social processes generally. In seeking to demonstrate how specific situated practices combine in-group interaction to constitute a small culture for that group, we orient towards Holliday's understanding of small culture because we regard it as helpful in avoiding an essentialist or at least over-generalised view of culture. Culture, in Holliday's small culture paradigm, is attached to 'small social groupings or activities wherever there is cohesive behaviour ' (1999, p. 237). A small culture does not bear a subservient or subordinate relationship to a large culture such as a national culture. Rather, small cultural formations exhibiting similar behaviours and practices can be found to cut across national, ethnic, and other large culture boundaries. With reference to our example, there are similarities in basketball practice in clubs worldwide. Most obviously, some of these can be attributed to the commonly understood rules of the game, and also potentially to globalised media in encouraging cultural behaviours and norms (see the third section). We can also regard the meaning-making and ultimately the learning that takes place in practice sessions and in matches themselves as opportunities for the emergence of cohesive cultural behaviour; such learning is not, of course, restricted to practice within one club. Hence, we can regard the specific embodied interactions of the basketball practice discussed below as group behaviour constitutive of a small culture, one which bears resemblance to other small cultures which have playing basketball as a goal.
This discussion is taken up again in the conclusion with reference to the behaviours and practices analysed in the body of this paper. In the next section, we describe the research setting in more detail, drawing attention to the ideologies of racism and discourses about a lack of social cohesion that circulate in political and media spheres, not just in the UK but across the global north. Both racism and concerns over social cohesion are commonly associated with contexts of superdiversity. Later, the microanalysis of video data from the training sessions is seen to counteract stories of unruliness by showing how social orderliness, cooperation, and creativity operate in the details of interaction. In the third section, we introduce basketball and its players, with a focus on how the sport's practitioners index the global and the local in their practice, and in their language, dress, and action. In the fourth section, we set out our theoretical framework and methodological approach to the analysis of the video data, collected as part of the broader ethnographic work at the site. The fifth section focuses on the microanalysis of that data, before we offer our conclusions in the final part of the article. These speak to the analysis itself; we also reflect on the how the subtleties of our visual ethnographic analysis within a translanguaging approach contribute to understanding how people's embodied communication contributes to the constitution of a small culture in a context of superdiversity.

The stigma of social disorderliness
The basketball training sessions take place in a community centre in the neighbourhood of Chapeltown, in inner-city Leeds. The origins of the centre are rooted in conflict and the resolution of conflict. In April and May 1981, Brixton, then Southall (London), Toxteth (Liverpool), Moss Side (Manchester), and Chapeltown (Leeds) were engulfed in what the media chose to call 'riots', in which shop windows were broken, premises fire-bombed and looted, and police and police stations attacked (Farrar, 1981). Whether these eventswhich in Leeds were initiated 'mainly but not exclusively by black youths' (Farrar, 2002, p. 231)were the result of purely criminal activity, reaction to poverty and deprivation, or the radical response of alienated and politically marginalised people to perceived injustice (Benyon, 1987, in Farrar, 2002 remains disputed, though it is probable that to some extent all these factors were involved. Whatever the case, in Leeds, the 'riots' had a significant impact on the Labour-dominated, left-leaning City Council, which over the next few years oversaw the construction of new housing in Chapeltown and neighbouring Harehills, as well as new health, training, and legal centres, and the extension and development of youth centres (Farrar, 2002, p. 237). It was around this time that the basketball club called the Chapeltown Warriors was formed. At the onset, as a result of its almost exclusively black membership, the club was subject to racist discourse. According to the coach in an interview: Fragment 1 (interview data) PE: there were stories about the Chapeltown Warriors that went around (.) you know (.) about players carrying knives (.) and tuhhh (.) you know (.) all sorts of-(.) that was just completely nonsense (.) but ((laughs)) but there were a lot of stories that went around.
Even today, multicultural Chapeltown continues to be stigmatised. This is reflected in the following interview extract published by BBC Leeds far more recently than the 1981 disturbances, in September 2014, in which a young man from Chapeltown speaks about where he lives.
A lot of people relate Chapeltown with violence and gangs, but in my opinion, it's no different to other inner city areas in Leeds […]. There's stigma attached to Chapeltown. People think they may get robbed and bad things may happen to them, which is not true. I'm not saying bad things don't happen but the probability I think is not any more than if you went to any other inner city area in Leeds or throughout the UK. (BBC Leeds, 2014, paras 2-3) Venturing into the community centre where the basketball training sessions are held, measures to counteract disorderly behaviour, real or imagined, are suggested in the linguistic landscape (Figure 1). The following photograph was taken in the reception area of the centre. The hand drawn sign on the mid-left reads 'STOP GUN CRIME, HELP SUPPORT PEACE', with the image of a fighter. To the right, we are warned that 'DOGS ARE NOT ALLOWED'. And in the background, Nelson Mandela looks on, presumably keeping a fragile peace.
In rooms at the centre devoted to work with local youth, signs prohibiting drugs, weapons, fighting, inappropriate dancing, and general aggressive behaviour, are also prominently positioned. It is not clear if the centre's management put up this signage in response to the behaviours it seeks to prohibit. It could equally have been placed as a pre-emptive measure, against assumptions of potential disorderliness.
Against this backdrop of assumptions of potential disorderliness, our ethnographic work highlighted how intricately ordered interaction emerged in the training sessions. The following reflections by team researcher Jolana Hanusova in her field notes from the first day she observed a basketball training session are illuminative in this regard: Fragment 2 (field note data) At this point I had already realized that this was going to be a much more structured and organized training than a game between friends that I was imagining. The playersthere was about 10 of them nowwere tying their shoelaces and chatting to each other, some of them were already at the court practising, each with their own ball.
Our work is concerned with communication in basketball in a context of superdiversity. Our study is part of a fast-growing body of work into translanguaging, which, as we explained above, encompasses the trans-semiotic, i.e. communication across multiple semiotic modes, and embodied communication. This being so, and although our analysis focuses mainly on visual rather than linguistic data, the following section will briefly introduce the spoken language and cultural aspects of basketball, as we encountered them in our fieldwork.

The language of basketball
Sport is not an area that has attracted much attention from language researchers interested in conditions of superdiversity. Lian Madsen's ethnographic study of young Taekwondofighters in Copenhagen (2008Copenhagen ( , 2015 is a notable exception. The taekwondo club is described by Madsen thus: The club forms a complex social space of Taekwondo traditions, Buddhist philosophy, and Danish leisure community culture combined with various cultural, social, linguistic and educational backgrounds. (Madsen, 2008, p. 199) The basketball team training sessions we describe in this article are likewise a 'complex social space', in which players from different linguistic, cultural, and social backgrounds meet to engage in focused activity, embedded within local and global dynamics. Madsen describes how practices within the taekwondo club are oriented to different scalar influences (local, city, state, and global). This is also the case for the basketball training sessions discussed here. Players and onlookers, most of whom live in the local area, dress in NBA kit, indexing the fact that despite the sport being played and watched across the world, the USA remains the pre-eminent centre of excellence, interest, and influence to which basketball coaches, players, and fans elsewhere orient (Jessop, 2012). In the USA, basketball continues to be associated with players of poorer ethnic minority background, and it is referred to as 'a black man's game' (Celzik, 2012; see also Lapchick & Guiao, 2015). As already noted, the basketball team we observed in Leeds originated in the 1980s as the only black team in the city league, made up of Afro-Caribbeans from the local neighbourhood. The team is still made up of players identifying as Afro-Caribbeans, along with African, Eastern European, and Filipino players. According to the team's coach (PE), basketball, and this team in particular which has no membership fees, attracts the disadvantaged: Fragment 3 (interview data) PE: there's lots of hoops in the parks and stuff like that (.) so you need a ball really to play basketball (.) you don't have to go out and buy a racket and special shoes and join a club and pay.
Thus, while in terms of status and economics the team are a world away from the glamour of the NBA, the link is strong in terms of the culture of basketball, the coaching, and play.
Basketball, like any sport, and indeed like most genres of human activity, has developed its own domain-specific terminology. However, as lexical items, the terms are seldom if ever exclusive to basketball, though within the confines of the game their meanings may be quite specific and distinctive. A simple categorisation of terms orients us to key features of the game, which we explore in more depth in our video analysis below. Basketball is characterised by narrowly defined goal-oriented activity which is both cooperative and competitive and performed within the affordances and constraints of both (i) strict though arbitrary spatial, temporal, and interpersonal frameworks, and (ii) an enframing which is not at all arbitrary, but directly linked to broader social processes and values. The basketball terminology that was noted in our ethnographic data, and that was used orally during the sessions, related to the following: . the playing area (the court) and its objectsmidcourt line, foul line, basket, hoop, backboard, three-quarter court, halfcourt; . players' actions in relation to the courtdrop, layup, run the floor, halfcourt defence, defend the corner, protect the middle; . players' actions in relation to space in generalpivot, jump; . players' actions in relation to the ballhold, pass, throw, dribble; . players' actions in relation to other playerscut, block, overplay, step inside, step up, step around, step across; . players' actions in relation to the whole playing environment (other players, the ball, the court)steal, pick, throw over the top, trap the ball, look (understood as an aspect of action ordering and communication, see Goldstein, 1994), and fake (as a form of intentional miscommunication); . players' actions in relation to timefast break, up the tempo, walk, run; . players' actions in relation to the physicality of interpersonal relationssqueeze, press, put pressure on, move and force, bump; . infringements of the rulesblocking, holding, pushing, charging, travelling, which have corresponding embodied symbols for use by the referee during a match.
Beyond the playing of the game itself (in coaching, for example), there are terms relating to attitude, fitness and conditioning, and skills (individual and team -'tactics'/ 'strategy', 'offense'/'defence'), and so on. There are also ways of speaking about the game ('play a bit of basketball', 'play ball', or just 'ball'), which point to the origins of basketball terminology in the USA. Indeed, in our data, we saw little adaptation of the language of global basketball culture in use at the scale of the local small culture developed through interaction in the club (e.g. 'offense' has not become the otherwise locally preferred term 'attack'). The discourse of basketball is also subject to wider influences, principally those generally understood as relating to Afro-American culture. These include music (e.g. hip-hop), clothing, hairstyle, and personal adornment, and capitalist discourses of financial reward, liberation from poverty, and making it big. Verbal communication is important to the development of the training sessions we observed amongst players and between them and their coach. Crucial though it is to the overall activity, it is not always the main means of organising the ongoing action. The list below sets out the phases that make up the basketball training sessions. The phases marked in parenthesis occur in only some of our data, and are thus considered optional elements of the training sessions. The episodes marked in italics are those in which speech, usually led by the coach, is more central to the organisation of the activity (i.e. phases in which long stretches of talk are heard). Other times, spoken language is used, but it tends to be mainly short exchanges between players, calls for the ball during play, counting during stretching, etc. Hence, in most of the training sessions, spoken language played a secondary role to other embodied forms of communication. With this in mind, in the following section, we present the methodological approach taken in analysing how multimodal communication emerges in the video data, before turning to our analysis and interpretations.

Methodology
As mentioned already, our basketball data collection took a structured visual linguistic ethnographic approach. Team researchers John Callaghan and Jolana Hanusova would observe weekly training sessions, write field notes, and record audio data, as well as making video recordings. Our analysis in this article builds on this ethnographic approach, though we give primacy to the analysis of video recordings. This analysis is influenced by ethnomethodology, which finds its roots in the work of Harold Garfinkel (e.g. 1964Garfinkel (e.g. , 1967. In particular, it is guided by Garfinkel's idea that ordinary people are not sociological fools but knowledgeable agents who draw on common-sense knowledge to normatively build intersubjectivity and work cooperatively with others to accomplish everyday goal-oriented activities. Ethnomethodological studies which are relevant to our own work on basketball include Evans and Fitzgerald's (2016) video analysis of basketball coaching sessions, which examined how participants locally negotiate the context of 'training' to make players' actions accountable. Haddington, Mondada, and Neville's (2013) work on the dynamics of language, embodied conduct, and spatial and material orientation, in interaction in mobile situations involving both micro moves (see below) and the movement of people's whole bodies from one position or location to another ('co-ordinated mobility'), also offers guidance.
In line with Garfinkel's ideas about the common-sense knowledge of social actors, we also draw on the notion of mental and bodily schemata (Anderson, 1977;Bartlett, 1932;Bourdieu & Wacquant, 1992;Piaget, 1923) that is, of patterned organisations of thought and/or action which categorise and relate experiential 'knowledge' and play a role in perception and interpretationand of the involvement of schemata in teaching, learning, and playing sport.
Finally, we draw on another collection of theories and constructs relating to the multiple modes of human interconnectedness, themselves strongly influenced by ethnomethodology. These include Scollon and Scollon's (2004) nexus analysis (NA) and Norris's (2003) multimodal interactional analysis (MIA).

Nexus analysis and MIA
Nexus analysis is a systematic and theoretically sophisticated form of ethnography designed to situate semiotic actions within their social and cultural processes and histories. It sees social life as constituted at the micro-level of social interaction, the level at which social categories and structures are enacted (Scollon & Scollon, 2007). NAand MIA, which applies the Scollon's theories to the study of multimodal datathus take the mediated action (Wertsch, 1998) as their focus and unit of analysis, and see action functioning at a range of 'levels' or 'scales' of time and place, from the micro to the macro (Lemke, 2000;Russell, 1912Russell, /1967Ryle, 1949). As anyone who analyses video data will quickly realise, action flows continuously, while analysis arrests to dissect, and heuristic devices, such levels of scale, are inevitably arbitrary in definition, number, boundary, and scope. Bearing in mind this arbitrariness, we have used a version of these heuristic scales in analysing the video data in this article, one adapted to the playing and pedagogy of the sport we are studying.

Heuristic scales
At the lowest level (micro level), sequential structures of social action are seen to be constituted by the smallest interactional units of meaningaudible in-breaths and out-breaths, u(h)ms and uhs, small bodily movements (including preparation, stroke, and retraction), and so on. Lower level actions or moves such as blocking, passing, and shooting are made up of multiplicities of enchained micro-level actions, and in turn enchain themselves to create mid-level actions or sequences of actions (such as those called plays in sports). Higher level actions (phases or episodes) created by actions at the levels below, ultimately cohere to constitute overarching actions or events, such as a game of basketball. This hierarchical relationship between scales is illustrated in Table 1.
Overarching actions have been seen by analysts working in a range of traditions as orienting frameworkssituation definitions, frames, scripts, schemata, genres, membership categories, inferential procedures, and so onwhich help define, for interactants and analysts, the nature of the action in course. This orientation involves linking concrete actions with 'abstract' templates existing on longer timescales (Lemke, 2000;Scollon, 2005), i.e. with multisensory procedural memories which synthesise previous actions of similar types or genre. Overarching actions are thus 'sustained both in the mind and in activity' (Goffman, 1974(Goffman, /1997 and are, according to some anthropologists, the principal units for encoding and transmitting cultural material (Blommaert, 2008, p. 2), which is what makes them such fruitful objects of study. In the course of our study, we came to see enchainments of actions at all levels above the micro as being organised by schematathat is, by associative networks of procedural memories. Since any given action, whether embodied or imagined, will involve the much same procedural memoryand, indeed, to some extent, the same brain cells (Eagleman, 2015) such a view of schemata allows us to go beyond the Cartesian dualism of mind-body (abstract-concrete), to jettison the linguistically biased idea of templates-for-action as scripts, and to work towards a unified theory of mind-body-world. Meanwhile, we take it that templates for action, the principal units for encoding and transmitting cultural material, are procedural memories of action; however, these are synthesised or reconfigured at various levels of scale.

Multimodality: 'modes' in interaction
In MIA, a communicative mode is seen as 'a heuristic unit that is loosely defined without clear or stringent boundaries and that often overlaps (heuristically speaking) with other communicative modes' (Norris, 2003, p. 101). That said, modes are culturally shared systems of representation (i.e. semiotic systems) 'with rules and regularities attached to them' (Norris, 2009, p. 79). As such, they are abstract 'grammars' (Kress & van Leeuwen, 2001, p. 22) realised by the use of material resources or media. The 'rules' of this grammar are dependent on use, rather than being determined by the medium (eyes, vocal chords) or the mode (gaze, speech) themselves. Approaches we draw on in our analysis, such as MIA, do not distinguish between medium and mode, using the term 'communicative mode' to encompass both. Moreover, 'there is no notion of a modal system outside of interaction' (Jewitt, 2009a, p. 34). Each mode (e.g. space, speech, embodied mobility in the video data below) in a multimodal ensemble is seen to realise different communicative work and to have been shaped for this purpose by its cultural, historical, and social use. People orchestrate meaning through their selection and configuration of modes (Jewitt, 2009b, p. 15) as semiotic resources. Meanings are 'shaped by the norms and rules operating at the moment of sign making, influenced by the motivations and interests of a sign maker in a specific context' (Jewitt, 2009b, pp. 15-16).
Consequently, any given mode is contingent upon fluid and dynamic resources of meaning, rather than static skill replication and use […] modes are constantly transformed by their users in response to the communicative needs of communities, institutions, and societies. New modes are created and existing modes transformed. (Jewitt, 2009b, p. 22) The conventions used for representing multimodal aspects of interaction in this study are adapted from different conventions found in ethnomethodological (and its sister discipline, conversation analysis) literature, NA, and MIA (e.g. Jefferson, 1984;Mondada, 2008;Norris, 2003;Scollon & Scollon, 2004), with screen shots from the video data being prioritised over symbolic descriptions of actions for ease of transcription and interpretation by the analyst and by the reader.

Research questions
These theoretical and methodological tools enable us to achieve the aim of the paper. As noted in the introduction, our work enriches the understanding of communication in spaces where languages, other communicative modes, and indeed cultures are in contact.
Informed by these perspectives, in approaching the data presented in this article, some general questions that we have been asking are: . How do individuals from diverse ethnic, linguistic, and cultural (etc.) backgrounds work together to produce orderly social action in basketball? . What role do language and the use of other embodied and disembodied (e.g. objects, space) resources play in coordinated action? . How is movement across semiotic modes transformative? . What underlying methods (in the ethnomethodological sense) for achieving intersubjectivity, or what 'web of practices that is so deeply rooted that it can transcend linguistic and cultural diversity' (Schegloff, 2007, p. xiii), are in play? That is, what part do universal or 'innate' practices play? . Which practices have to be learnt/taught? How are they taught/learnt? . What are the (informal) pedagogical implications in contexts of superdiversity?
Bearing in mind the explicit focus on cultural activity which we adopt in this paper, we also ask how the transformative practice of trans-semiotic interaction can be understood as contributing to the development of a small culture.
With such questions in mind, we turn to the analysis of some of the video data collected during our ethnographic fieldwork. As mentioned earlier, this paper reports on the phase of the TLang project concentrating on interaction in sport contexts. The Sport case study in Leeds followed our Key Participant Tiago from Mozambique and his life-shaping involvement in two activities, capoeira (not discussed in this paper) and basketball. Data were collected between September and December 2015. They comprised: observation field notes (22 sets, including 9 of basketball, 43,652 words); audio recordings (13:42 hours in sport settings, transcribed where intelligible and audible, and 4:36:42 hours in Tiago's home); video recordings of sport practice (5:30:57 hours); and 13 informal and more formal interviews with Tiago and those around him (12:23:52 hours). Data for this paper come mainly from video-recorded basketball training sessions, filmed using cameras fixed at each end of the court, and audio-recorded using a voice-recorder attached to Tiago's chest. Informed consent was gained from all participants to make the recordings and use them for our research purposes. The researchers followed the ethical protocol developed for the TLang project as a whole, scrutinised and approved by the University of Birmingham's ethical review committee. Initial work on the video data was carried out using the multimodal discourse analysis tool ELAN, aiding the identification and preparation for analysis of the higher-, mid-, low-and micro-level actions of an event as sketched out in Table 1.

Analysis: coordinated action, communication, and creativity
This analysis begins with a first data extract from a warm-up phase from a training session that serves as a first counter-argument to those claiming social disorderliness in superdiverse Chapeltown and at the community centre where our research was conducted. The analysis continues to develop our argument about the intricately coordinated and also creative nature of the basketball training sessions by introducing the notion of schemata, or of patterned organisations of thought and/or action, in relation to the phases of coaching and playing basketball in sessions.

Counter-argument 1
In analysing the basketball data at a micro level, we have focused on how bodies coordinate with speech and other semiotic resources, in extremely orderly ways, in organising moves, plays and phases, including the transitions between different sequential structures of social action that make up the basketball-training event. The following excerpt is a telling example of how speech and bodies coordinate in time and space in accomplishing one such transition.
The excerpt takes place at the point of changeover between a stretching activity, led by one of the players, and a shooting drill, for which the coach gives instructions (i.e. between phases 4 and 5 in the list in the third section). The session was recorded using two cameras placed at each end of the court, although the view from only one camera is included here. The players are positioned around the walls of the court, and are stretching to the count from 1 to 10 by one of the players, who is acting as leader. The coach paces away from the wall. The player's (PL in the transcript) counting has been transcribed, with pauses (in tenths of a second) marked in brackets, and the image next to each utterance is a screen shot taken at the precise moment when that utterance begins. The images reveal how the coach (PE in the transcript) takes a step towards the centre of the court each time a number is called out. The entire fragment lasts just under 15 seconds (see the start and end time of the excerpt).
This very short extract highlights at least two aspects that are of interest to our analysis. Firstly, it shows how despite not being an active stretcher in the stretching episode of the training session, the coach (PE) synchronises his walking activity with the players, by taking a step with each new number. Secondly, the extract shows how this orderly walking is also goal-driven, in the sense that it is oriented to arriving at the centre of the court, at precisely the sequential moment (the kairos, Erickson, 2004) at which the stretching episode is due to end and the coach needs to give instructions for the next episode. As a walking rather than a stretching subject, we argue, the coach is nevertheless an important participant in the organisation of the activity, in a similar way to how Goodwin and Goodwin's (2004) listening subjects were important to the maintenance of oral interaction. Through the intricate coordination of speech, bodies, space, and talk semiotic resources that participants mobilise and interpret in achieving intersubjectivity moment by momentthe structural organisation of the event emerges in a highly orderly and cooperative manner. Such orderliness, we argue, is quite a distant reality from that of the violence and antisocial behaviours suggested by external accounts of Chapeltown and the community centre and by its internal linguistic landscape. Such orderly action is a key feature of the remainder of the data presented in this analysis.

Counter-argument 2
The above example highlights the probably unconscious synchronisation of different kinds of actions in a basketball stretching exerciseunderlying which, perhaps, is that 'web of practices' of which Schegloff speaks, 'that is so deeply rooted that it can transcend linguistic and cultural diversity' (Schegloff, 2007, p. xiii). However, there is also ample evidence in our data of orderly practices which have to be learnt/taught, and while some of these come into the categories of micro-actions and moves (throwing, catching, and 'laying up'), many take the form of 'generic orders of organisation' (Schegloff, 2007) those building blocks of social life which help orient social actors and allow them to organise lower level actions into meaningful and productive social events. Such mental-bodily schemata, functioning as templates for conduct, thoughts, feelings, perceptions, judgements, and so on (and as the actual thoughts, feelings, etc.) derive from the collective representations of the group which transmits them (Bourdieu & Wacquant, 1992, p. 7). What makes sport such a fruitful field of study for these kinds of organising frames is that their transmissionand indeed analysis by practitioners, coaches, pundits, and others is purposefully explicit. Thus, ongoing evolution of schemata (driven by the creativity and/or ignorance of participants, or by hybridisation resulting from diverse histories of practice, etc.) is often in plain view. In what follows, we present an example of how one kind of schema, a defensive formation in basketball, is introduced into novices' repertoires. This illustration incidentally highlights the flexible and contingent nature of schemata as adaptable templates (as opposed to fixed scripts) offering, at each juncture, a range of options for relevant action. Following this, we go on to sketch in some of the mechanisms which drive evolution in schemata.

Schematising action
Gathering the players together in what is neither the changing area nor the basketball court but a liminal space dedicated to preliminary instruction (Figure 2), the coach explains the defensive system he wants his players to use in an upcoming league game. To help them visualise the system or schema, he uses a magnetic board holding blue and white discs. In the transcription, the coach is moving counters around, pointing, or indicating actions on the magnetic board during most of the silences and speech, although we have not included these images due to space restrictions.
Fragment 5 (video data) PE: alright (0.8) okay. (1.3) the:: defence that I'm gonna want us to work on next week (1.1) the one-three-one↑ (3.0) but we're gonna play (.) effectively three-quarter court (.) one-threeone to put some pressure on the ball. (.) this is wherever the ball is. (0.8) these two guys are gonna come and trap the ball. (0.6) we're lookin' to trap the ball (.) t'make him throw that pass over the top. (.) this guy steps up. (.) we're lookin' to squeeze wherever we can. (.) so we're lookin' to either make him (.) pass that ball (.) try and pass the ball across so we can look for the steal (.) or (0.5) trap him in the trap him at half court so he's got nowhere to go. (.) s'even better to trap him this side of half court so he can't step back. (1.3) alright?
Following this verbal and graphic exposition, the coach gets five of his players (Team A) to take up positions on the court in the particular defensive formation (one-three-one) called for by the schema, thus providing a mental, embodied, and spatial experience of the schema's requirements. Other players (Team B) look on (Figure 3), plausibly learning through by putting themselves in the place of the players in Team A.
Dispensing with his magnetic board, the coach now takes up a ball, modelling the actions of an attacking player, thus giving his charges chance to perform defensive actions which they have already mastered to a greater or lesser degree. The schema is thus seen to enable players to enchain and organise existing lower level skills into more effective action during a period of non-possession of the ball (Figure 4).
The coach now takes up a variety of positions (as suggested by an orthodox offensive schema) from which to model passing the ball, and provides verbal instruction to his players on their most advantageous responses in terms of actions and positioning. In other words, he lets them experience the various options provided for by the flexible and contingent nature of this schematically ordered mid-level action (Figures 5 and 6).
Following this, the coach invites Team B onto the court to take up positions according to a previously learnt offensive schema and attack the basket defended by Team A. This allows Team A to put into practice the new defensive schema (one-three-one) which they have just been introduced to, this time in more realistic and demanding conditions (Figure 7). From time to time, the coach stops play and comments on satisfactory and problematic aspects of the players' realisation of the schema (Figure 8).
Finally, the coach brings the exercise to an end ( Figure 9) and provides some final advice on incorporating the schema into the higher level schema of defensive play and that of the game as a whole, relating this to a 'real' and, for the players, high-stakes game to be played at the weekend. Thus, we see the complete trajectory of the transmission of a mid-level action, though it will be some time before the flow of its constituent actions and the decision-making which gives rise to them become automatic, unconscious, and cognitively effortless as the practices become 'hard-wired' in what used to be thought of as 'muscle memory' but is now known to be neural tissue, first in the brain, then later in the spinal cord (Eagleman, 2015).

Anti-schema: disrupting schematised action
The effect of schemata is, among other things, to enable social actors to identify events and choose, consciously or unconsciously, from a range of appropriate actions, all the while anticipating and responding to the actions of other participants who are orienting to the same schemata. Because schemata make available shared patterns for the coproduction of action, underlying their smooth functioning is the principle of cooperation. Schemata rely on and facilitate teamwork. In sport, however, though opposing teams must cooperate in orienting to a master schema in order to produce a game (at a given time and place and following given rules), within the schema of the game itself, opposing teams are motivated by competition; to be successful, a team's competitive action must be performed with either greater speed or greater power than one's opponents', or they must introduce the creative element of surprise or deception. Gameplay, particularly offensive gameplay,  calls for the unexpected, which means something from outside the prevailing schema, thus something which will disrupt it. Figure 10 provides an example of such creative, disruptive, anti-schematic action.
The role of gaze in organising social interaction has been extensively studied (Argyle & Cook, 1976;Argyle & Dean, 1965;Exline & Fehr, 1982;Goodwin, 1980Goodwin, , 1994Goodwin, , 1995Kendon, 1967Kendon, , 1978. However, it requires little reflection to conclude that under normal circumstances, the direction of an actor's gaze indicates (i) their intended direction of their action and (ii) where relevant, the intended 'recipient' of the action. Here in Figure 10, however, we see a basketball player (on offense) exploiting the expectations normally triggered by direction of gaze to confuse his opponents and thus increase his chances of successfully completing a pass. 'Giving the eyes' one way (i.e. providing a visual miscue) and throwing the ball another (doing the unexpected), the player to the left anticipates the likely responses of his opponents in order to subvert them. He thus demonstrates that to creatively disrupt a schemapurposefully, at leastone must first have mastered it and its constituent actions.  No matter how long a sport has been played, there is always, it seems, the possibility of the introduction of new and unexpected action. We have only to think of the 'Fosbury flop' in high jumping, the 'slam dunk' in basketball, the 'reverse sweep' and 'reverse swing' in cricket, the 'Cruyff turn' and 'sweeper-keeper' in football, Chris Froome's 'super-tuck' in the 2016 Tour de France, and the Italian Rugby Union team's 'anti-rucking' strategy seen recently and for the first time in the international game's 146-year history. Such anti-schematic actions, functioning at various scales from micro-actions to higher level strategic levels, may be the products of imagination and creativity as well as of hybridisation of practice resulting from global processes, including the increasing mobility of people and ideas, and the influence of global media.
Super-schema: embracing and responding to the anti-schematic action Since purposeful anti-schematic action depends for its success on surprise, it can only be performed a certain number of times before it gains the status of expected action and gives rise to effective counter-action or negationin other words, before it becomes absorbed into the schema. This embracing of the anti-schematic action results in a more highly evolved schema, what we may call a 'super-schema', and is an ongoing process in the life of a sport, and social life in general. However, while this embracing of a schematically disruptive action goes some way to countering it, it cannot always entirely negate the action, since the performer now has two viable options. For example, in basketball, an offensive player may now throw the ball in a direction other than that of his gaze, or throw it in the direction of his gaze, as per the original schema. The new, super-schema may embrace both these possibilities, but if the initiating actor alternates randomly between his options, his opponents will struggle to anticipate his actions, even though they know what these options are. This is why, in Fragment 6, illustrated in Figure 11, Tiago's (TI) team mate, Anderson (AN), exhorts him to 'mix it up'in other words, to exploit the possibilities of unexpectedness from within expected options. In passing, we should note that this incident also illustrates the ongoing transmission of schema, this time during the course of a practice match, and by a more expert teammate rather than by the coach.

Discussion and conclusions
In this paper, we introduced the complex social space of basketball training sessions in which players from different backgrounds meet to engage in focused activity, embedded within local and global dynamics. We set out by situating the neighbourhood, the community centre where our fieldwork was conducted, and the basketball team itself in relation to stigmatising discourses that suggest an environment of social disorderliness. Such discourses, we argued, are frequently associated with superdiverse contexts such as the one we study.
Returning to our research questions, we sought to understand how individuals from diverse backgrounds work together to produce goal-oriented action in the basketball training sessions. Our microanalysis of video data counteracted the stories of unruliness in conditions of superdiversity by showing how social orderliness, cooperation, and creativity unfold in sophisticated ways in the details of interaction. We focused mainly on embodied action, given its predominance in the data over spoken interaction, and its general salience in the sport setting. We highlighted deeply rooted webs of practices that appear to transcend linguistic and cultural diversity in the basketball training sessions (e.g. a transition from stretching to drilling), as well as how new practices or shared schemata are taught, learned, and creatively deployed (e.g. a defensive formation). We noted how, for example, due to the competitive aspect of sport, prevailing schemata are often disrupted for goaldirected purposes, but that schemata have a tendency to embrace the disruptive in orderly ways and that this is a key mechanism in their evolution. Anti-schematic actions, functioning at various scales from micro-actions to higher level strategies levels, may be the products of imagination and creativity or, in contexts of superdiversity, products also of the hybridisation of practice resulting from global processes, such as the mobility of people and ideas, and the influence of global media.
Because of the particularly goal-oriented nature of sports, they are fruitful fields of enquiry for the study of how communication unfolds in contexts of superdiversity in orderly ways. This contributes to our understanding of translanguaging, the overarching conceptual frame for the project within which this study is embedded, as extending beyond language, to encompass multiple modes of communication and interconnectedness. Visual linguistic ethnographic research attention on sport settings, and the crucial role in communication in such settings of the visual and of embodied action, serves to deepen this understanding.
Moreover, the orderliness, cohesive behaviour, and collectively learnt practices of the goal-oriented action in the training sessions that we observe contribute to constituting the cohesive cultural activity of the club itself. While the activity in the basketball club involves social actors from varied backgrounds, their activity in practice sessions and in matches combines to create its own small culture (Holliday, 1999), developed in unfolding cohesive, unified, goal-oriented verbal and embodied interaction. The orderly practices we observe that unfold in the basketball training sessions becomeas feminist economic geographers Katherine Gibson and Julie Graham (2008) put it -'"real", more credible, more viable as objects of policy and activism, more present as everyday realities that touch our lives and dynamically shape our futures ' (2008, p. 618). In doing this, we aim to contribute to what Gibson-Graham refer to as the 'performative ontological project' of reinscribing meanings onto our world.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
The research was funded by the UK Arts and Humanities Research Council [grant number AH/ L007096/1] as part of the project 'Translation and Translanguaging: Investigating linguistic and cultural transformations in superdiverse wards in four UK cities' (TLANG) led by Angela Creese (University of Birmingham).

Notes on contributors
John Callaghan is a researcher on the TLANG project at the University of Leeds. His interests include linguistic and visual ethnography, in particular when used to investigate intercultural interaction in relation to its spatial and historical contexts.
Emilee Moore is Serra Húnter Fellow at the Universitat Autònoma de Barcelona and a postdoctoral researcher in the Beatriu de Pinós programme (Generalitat de Catalunya), affiliated with the TLANG project at the University of Leeds. She studies interactional practices in multilingual and multicultural contexts from a perspective integrating linguistic ethnography, interactional sociolinguistics, ethnomethodology, and sociocultural theories. James Simpson is a Senior Lecturer in Language Education at the University of Leeds, and is a coinvestigator on the TLANG project. In his research, he specialises in communication in multilingual settings, the teaching and learning of languages in migration contexts and the intersection of new technology, literacy, mobility, and social justice.