“ I just sit, drink and go back to work . ” Topographies of language practice at work

: The paper explores the in situ negotiation of in/exclusion in and through language in a multilingual professional setting, paying special attention to the relationship between language and space. We argue that multilingual practices and material space are co-constitutive; individuals enact group membership and professional roles spatiolinguistically and re/produce in/visible social and material boundaries. Despite the well-established literature on in/exclusion, the ways in which it is negotiated in asymmetrical, emplaced, workplace encounters is still underexplored. We introduce a topographies of practice framework and show how professional asymmetries are enacted in and through language choice and language use in the multilingual workplace. We take an Interactional Sociolinguistic approach and report on the analysis of 23 h of interactional data and 42 h of ethnographic observations from a professional, multilingual kitchen in Finland. We show patterns that are un/marked in the data and constitute the norms in this particular workplace. We argue that topographies of practice are topographies of in/exclusion enacted in and through situated encounters; we pay special attention to the role of employees who are legitimised to cross visible and invisible boundaries and we close the paper with recommendations for future research.

has continued addressing the complexities associated with multinational organisations where linguistically diverse and highly competitive environments are the norm. The aim of this paper is to expand earlier research by focusing on multilingual practices, specifically in relation to the employees' material space, drawing on the case of a multilingual professional kitchen.
Sociolinguistic and Applied Linguistic research have repeatedly shown that language choice and use at work is ideological and has direct implications to employees' accessing (or not) capital (as defined by Bourdieu 1986). This is done in formal and informal encounters which are spatiotemporally situated. We zoom into this process and look at the ways in which boundaries are constructed and policed by (dominant) language speakers (Finnish in our case) in daily practice at work. We show how this is done in the linguistic moment and reflect on the implications for understanding multilingual practice in the workplace and in society more broadly. We situate the work in Workplace Sociolinguistics and the Interactional Sociolinguistic (IS) tradition, following the paradigm set by Gumperz (1982). Although Workplace Sociolinguists have engaged with space in multiple guises throughout the history of the field, detailed analyses of employees' practices positioned in a material space have not been the focus of our enquiry until recently.
In more detail, workplaces are political spaces (Clegg et al. 2018), with hierarchies that are negotiated in situated language practice and in/formal work areas. Materiality in workplaces is intertwined with social processes and is an integral part of interaction (e.g., Leonardi 2012) and meaning-making (Boxenbaum et al. 2018).
Work and workplaces are central to the social order. Accessing and securing work as well as career progression are critical for economic growth, social mobility, mental and physical wellbeing. Particularly for migrants and those coming new to a socioeconomic environment, the workplace is where settlement and 'integration', a contested term, is actively negotiated. Classic research (e.g., Gumperz 1964Gumperz , 1982 and more recent sociolinguistic studies (Roberts 2021) have shown the power asymmetries between the established 'host' community and those who need to claim being 'one of us'. This line of research has provided ample evidence on the cost of deviating from the dominant or standard. The term linguistic penalty, the price those who do not 'fit' the majority pay, is well established in the disciplinary jargon and recent studies (Kirilova 2017;Kirilova and Angouri 2018;Roberts 2013) continue to provide evidence for the politics of gatekeeping at work. The IS tradition has provided the main theoretical and methodological approach for this body of work.
We explore here negotiations of in/exclusion in the material space of a professional kitchen and show the impact on the daily experience of employees in this setting, as echoed in the quote that forms part of the title of this paper. The professional kitchen provides an example of a complex work setting, well-described by Pennycook and Otsuji (2014: 175) as "a spatiotemporal hub crisscrossed by trajectories of people (cooks, floor staff, phone calls), artefacts (knives, sieves, plates) and food (ingredients, cooking, finished items)." Modern kitchens are also a prime example of the labour market and "wages, skills and visas that bring many backgrounds" together (Pennycook and Otsuji 2015: 70). As such, it constitutes an empirical context worthy of study.
The paper is structured in four parts; we first turn to the discussion of language and space, before moving to the description of the methodology and analysis. We close the paper by revisiting a topographies of practice framework for the study of the intertwined relationship between language and space and language practice, and suggest directions for future research.

Spatiolinguistic practices in the workplace
Space is commonly understood as lived, political and actively produced, after Lefebvre's (1991Lefebvre's ( [1974) influential work which has deeply marked thinking in our discipline and his triptych of 'perceived', 'conceived' and 'lived' space. According to Lefebvre practices shape and are shaped by material space. In his words: Spatial practice […] simultaneously defines: places ─ the relationship of local and global; the representation of that relationship; actions and signs; the trivialized spaces of everyday life; and, in opposition to these last, spaces made special by symbolic means as desirable or undesirable, benevolent or malevolent, sanctioned or forbidden to particular groups (Lefebvre 1991(Lefebvre [1974: 288).
The relationship and distinction between space and place has been left underexplored until recently and it is common for the two to be used interchangeably. The general consensus is for 'place' to refer to specific and enduring locations while 'space' is seen as a process, relationally negotiated and fluid (to varying degrees) (Pennycook and Otsuji 2015). The distinction between the terms, however, is largely open to the interpretation of the researcher. We are concerned here with enduring patterns in the context of one professional setting and the relationship with recurrent language practice. 1 Places are not 'containers distinct from their contents' (Lefebvre 1991(Lefebvre [1974: 87). Employees form groups in the multilingual workplace around roles, languages and other shared characteristics (see e.g., Humonen 2019 on non-Finnish speaking migrant employees; Kim 2018 on Korean expatriates in the UK; Kingsley 2013 on multilingual and transnational banks in Luxembourg). Although these studies did not take a spatial approach per se, they have shown that groups have a spatially negotiated footprint and boundaries that are actively guarded and/or challenged and redistributed. Those findings align well with the re/ignited interest on space and language and the aspiration of Workplace Sociolinguists to study groups interacting in their professional settings holistically.
Further on this, space has been central in Sociolinguistic studies since the very beginning of the field. Variationist Sociolinguistic studies, in particular, have explicitly connected space with language use since Labov's seminal work in Martha's Vineyard island and New York City stores (Labov 2006(Labov [1966). The relationship between geographical distance and language change, national and regional borders and language contact, and the geo-and social-space of speech communities have been prominent on Variationist Sociolinguistic studies (for the work of core variationists e.g., Britain 2010;Eckert 2012). Space, according to this agenda, constitutes a synthesis of geographical, social and contextual factors since early days of the discipline; classic work (e.g., Trudgill 1974) already showed that social stratification involves linguistic and spatial considerations.
Workplace Sociolinguistics also maintained a close interest in material space; this shifted from a narrow understanding of space, as the site of research, to a focus on the place where social order is negotiated and powerful centres and peripheries are perpetuated or resisted. The field turned to multilingual practices in institutional and corporate workplace and addressed the diverse linguistic ecosystem that is the norm across different geographical localities and linguistic contexts; mobility, migration and integration (and lack of) in the labour market are common themes of special issues and volumes (Vine 2017; for an applied linguistic approach see Canagarajah 2020). Research addressing boundaries and boundary crossing ) has expanded earlier scholarship on centres and peripheries. This work has shown: (1) the trajectories to and from linguistic and geographical centres to multiple peripheries and back again, and (2) the opportunities and cost individuals pay being constrained to remain at the periphery (cf. Wenger's 1998 notion of legitimate peripheral participation) and the barriers that keep them there (Canagarajah 2020;Holmes et al. 2011).
Although Workplace Sociolinguists engaged with space through the different panels of the kaleidoscope described above, there has been little work bringing together the material space and interactional order. Current work attempts to re/address this balance and looks into the dialectical relationship between practices situated in space and place and the meanings they mobilise, carry and perpetuate (Kusmierczyk 2013;Mesinioti 2022). This turn has also methodological implications favouring multimodal approaches replacing the focus on audiolinguistic data and moving from studying interaction in 'fixity' (see Britain 2016, on the limitations of 'sedentarism' in sociolinguistic research) to spatiolinguistic 'fluidity'. The IS approach, which is widely adopted by scholars in the field, allows for the combining of micro-level interactive processes with the wider material and non-material organisational setting. We elaborate further in the next section and introduce the framework we propose in this paper.
2.1 Topographies of practice at work: an IS informed framework for the study of language and space IS shares with conversation analysts (CA) the focus on situated encounters. CA has provided evidence about the interactionally negotiated relevance of space since the founding work of 70s (e.g., see the influential Schegloff 1972 paper on place formulation) which expanded the thinking of pioneering work on the interaction order (Goffman 1983). Recent CA work (e.g., Angouri and Mondada 2017;Mondada 2011Mondada , 2017 has turned to the material environment of interaction and drawing on work by Goodwin on embodied participation (Goodwin and Goodwin 2004) broadened the study of embodied actions to include the spatial distribution of encounters. Research on interactional space through a multimodal approach (e.g., Mondada 2009Mondada , 2012 has shown the systematic interactional accomplishment of space as a process situated in the material environment of the interactants. This develops the methodological inventory for the analysis of interaction and at the same time brings fluidity in the conceptualisation of space which is in line with the thinking in other parts of the discipline, notably in Applied Linguistics where a growing attention to spatial repertoires reflects the same trends (Canagarajah 2019). Canagarajah (2019) provides a useful discussion of the use of the term drawing on Pennycook and Otsuji (2015), and suggests that '[a repertoire] refers to the mix of codes, gestures, objects, and environmental resources that enable one to accomplish a communicative activity' (Canagarajah 2019: 11). The repertoire concept, originally used by Gumperz to refer to 'the totality of linguistic forms regularly employed in the course of socially significant interaction' (Gumperz 1964: 137), is being re/defined and expanded in studies taking an ethnographic approach to connect the individual experience to the material space and the use of linguistic resources. This provides an opportunity for new empirical insights and methodological and theoretical advancement.
Against this backdrop, we introduce the term topography of language practice to build on the situatedness of interaction in a sociomaterial and (small p) political workplace and propose an analytical framework for the socially embedded study of organisational language practice. We argue that the topography concept enables us to turn our attention to the material space and the process of crossing spatial and linguistic boundaries that challenge or perpetuate the spatiolinguistic status quo in interaction. We see language and materiality as co-constitutive and we are particularly interested in enduring spatial patterns of interaction. The term topography as per the Greek definition of the noun τοπογραφία signifies the analysis of the features of a structured ensemble. The term is preferred as it "stresses the unity of landscape components […] and embodies two potent geographical ideas that are applicable to all components of landscapes -place and space" (Huggett and Cheesman 2002: 3). As such, it constitutes a useful analytical metaterm enabling spatiolinguistic description of recurrent patterns and analytic frame for the practices negotiated in a situated encounter -constituting and constituted by the pre-existing structure of the social order (Angouri in prep).
The social/spatial order pre-exists the individual and is re-enacted or resisted in the situated interactional moment. By extension, local (language) ideologies are both emergent and engrained in the institutional order and material context of the employees as well as related to resources available to negotiate roles in the workplace. Together this forms an inseparable and continuously renewed whole (see Figure 1) which is enacted in situ and speaks directly to the broader sociopolitical order and hegemonic ideologies at societal level. In/exclusion is not a linear, static binary in this process. Rather, it negotiated in and through the social and local meaning of the multiple cues the interactants mobilise and negotiate in the asymmetrical context of a multilingual workplace. The relationship between those dimensions is summarised in Figure 1.
The framework brings together the material and linguistic 'here and now' and dominant (language) ideologies. A topographical view of language practice enables us to explore how workplace material space is enacted and delineated through recurrent daily practices and in relation to forming group membership and professional roles at work. We show how the model can be implemented in Sociolinguistic Workplace research in the light of the data below.

Methodology and research context
Our dataset is drawn from an ethnographic project on multilingualism and gatekeeping, conducted between 2017 and 2018, at a Finnish multinational food corporation, "Tasty Co". The corporation employs nearly 15,000 employees in its different business areas and subsidiaries, however, the focus of our study is specifically on the restaurant business division, pseudonymised as "Dining and Catering Services" (D&C Services). The overall data comprises 10 research sites across various organisational levels, out of which seven were restaurants. For this paper we zoom in to one multilingual restaurant named "Finlicious". Analysis of this site illustrates the relevance between fixity and fluidity that a topographic reading enabled us to offer. By focusing on one restaurant, our aim is to provide an in-depth description of the topographies taking place within this specific context between the restaurant staff.
During the fieldwork Finlicious's kitchen team consisted of 11 workers and depending on the day additional 6-8 service employees. The analysis draws on 23 h of interactional data, 42 h of ethnographic observations and extensive fieldnotes. As participant observer the second author was working side by side with the kitchen team for 5 days. This allowed for an opportunity to experience first-hand the daily work practices and socialisation activities in the context of the employees. Familiarisation with the material space of Finlicious and the different employee groups allowed for ongoing observation of emplaced practice even though space was not the focus of our initial project. This was then followed by collection of recordings and analysis of interaction. The audio-recorded data was collected by first placing the recorders on kitchen shelves and then by the researcher and a voluntary participant carrying the devices. We found the latter approach to be better suited for collecting interactionally meaningful data, as leaving the recorders on kitchen shelves led to the data being too scattered for analytical purposes due to high noise levels and the constant movement of the employees.
Negotiating access and ethical clearance is a significant part of any workplace research project. Once the study was authorised by the senior management, the final research proposal was presented by our gatekeeper to a group of restaurant managers with the aim to invite volunteers to take part in the study (for further detail on the rationale of participatory IS designs see Angouri 2018). The second author discussed with core stakeholders the characteristics of suitable research contexts for the study but did not take part in the site selection process. Our main gatekeeper emailed the schedule and locations of the restaurants to the second author. Because of the power differences between our gatekeeper and the restaurant managers, we took extra measures to ensure our participants became involved on a voluntary basis. The second author called each participating restaurant manager prior to the fieldwork in order to confirm consent, to provide further information on the project and/or an opportunity to reschedule or withdraw from the study. The research project was (re)introduced to all participating teams in morning briefings and written consent forms were reconfirmed with each employee verbally every day prior to recording.
In interpreting the data, we took a social constructionist approach and acknowledge the subjective nature of reality. The subjectivities (see Sarangi 2007 for the research paradox in workplace discourse studies) and the researcher's positioning arising from doing fieldwork were revisited recurrently by the authors (e.g., see Humonen andAngouri 2022, 2023). Following the principles of IS, we analysed the data using a model introduced in Workplace Sociolinguistic literature (Angouri 2018) enabling to connect the micro moment of interaction with the wider institutional order drawing on patterns in the data. This involved an analysis of features that are recurrent in the data as well as those that are marked and their consequences, suggested by the uptake and the sequential design. Building on this work, the topographies of practice framework provides a further step towards capturing the dynamic relationship between language and material space.
The main focus in this paper is on the kitchen staff (see Appendix 2 for a brief overview of the team members). We start the discussion of our findings with positioning the kitchen employees in their material space and provide a topography of their workplace.

The kitchen topography
Tasty Co's official language at the headquarters is British (specifically Oxford) English, as stated in the corporation's language policy and guideline documents. Unlike the parent company, D&C Services does not have written language policy/ guidelines. However, Finnish plays a central role in formal communication and recruitment. 'Excellent Finnish language skills' are listed as a prerequisite for all kitchen positionsincluding jobs that require little to no interaction with others/ customers, such as in cleaning and dishwashing jobs. Despite this language requirement, in practice, many lower level kitchen positions are filled with migrant employees who tend to use English at work (Humonen 2019). Hence Finnish and English co-exist in the context of the suitably ambiguous formal/informal language policy and are skilfully used to negotiate boundaries between groups and individuals.
'Finlicious' represents one of the busiest restaurants of the D&C Services Group, with the capacity of serving over 700 diners. It is located in one of Helsinki's central business districts and the majority of customers are English speaking expatriates from nearby multinational enterprises. On first sight then, it appears to be a contemporary international workplace where multilingual practices have become the norm. We will show, however, that the kitchen, backstage, topography emerges through the emplaced negotiation of dominant local (language) ideologies which perpetuates the in/visible social and spatial boundaries between the Finnish and non-Finnish speaking kitchen staff.
Movement in the kitchen space is part of carrying out the activities associated with the core business goals, and employees' emplaced positions are closely connected with their professional roles. Access to the organisational languages (or not) presents an extra layer of complexity to the workplace ecosystem. Previous research on professional kitchens has shown the consistency between language ideologies, hierarchy and the production of space. Work on the latter focused on the construction of authenticity of ethnic or local cuisines (e.g., Abas 2019; Maegaard and Karrebaek 2019). In addition, aspects of asymmetrical division of labour, with ensuing difference in reward, between front and backstage staff, migrant/ non-migrant status workers have also been addressed (e.g., Barrett 2006;Beriss and Sutton 2007). The kitchen space is not an equal space. Even though kitchens are inhabited by waiting staff, management, dishwashers, chefs and cooks, working in such professional environment normally include adherence to the strict hierarchy (Burrow et al. 2015;Murray-Gibbons and Gibbons 2007), with the Chef and Sous Chef in charge. The spatial layout of the kitchen and workstations reflect employees rankings and has an impact on who can say what and when. As such, it is a particularly suitable work context for studying topographies of practice and the ritualised activities that re-enact or challenge invisible borders. This process is subject to a fine balancing act between agency and top-down constraints as we show in the data (and Figure 1). Kitchen staff do not mark their use of space by personal objects (compare with the 'prototypical' white-collar office space) and there is a constant flow of activity which signifies working hours. The map below ( Figure 2) provides a representation of the kitchen topography based on the researchers' reading of the analysis of the data. It seeks to allow the reader to position the ensuing analysis in the spatiolinguistic ecosystem of this workplace.
The hot line is the "heart of the kitchen" (Demetry 2013: 584), which is managed by the Chef. Usually this is the most visible position both physically and symbolically. Whereas cooks typically specialise in the specific aspects of food preparation such as baking, grilling or cold dishes etc., the Chef is expected to do all of the above alongside keeping track of the budget and managing the staff. However, Finlicious's Executive Chef, "Pekka", is a newcomer who has been in the restaurant for a couple of months only. He is rarely seen in the kitchen as he spends most of his work hours in the office with the restaurant manager, "Ulla", learning about the inventory, food orders, etc. Pekka and Ulla have a monitoring role as part of their institutional authority. In Figure 2 we show the positions of our core stakeholders in the material space, corresponding to the base layer of Figure 1 theoretical framing (see also Appendix 2 for contextual information of the participants and their linguistic profile, note the L1 and non-L1/English/Finnish distinction in particular). The Chefs, Pekka and Heikki, are located in the centre of the kitchen from where they can observe the employees. Their central positionality is not random. Recent work on teamwork practice (Mesinioti et al. 2019) has shown how movement into material spaces and specific positions from where the whole encounter is visible, is part of enacting a senior role in a workplace context. The same applies here.
The culinary industry has been consistently reported to have normalised aggression among kitchen staff (e.g., Meloury and Signal 2014) and this is also the image portrayed in the media (e.g., Gordon Ramsay's Hell's Kitchen programme). However, as Finlicious serves mainly buffet style lunch and accepts ordered meals only at the grill station, the role of the Head Chef is different from restaurants with table service; namely, Heikki is not responsible for dealing with a rush of tickets (i.e., food orders), instead, his role mainly involves ensuring the overall operations of the kitchen; e.g., briefing the staff on the daily menu, guiding and helping with the food preparations, assigning different tasks, doing inventory, etc. As such, despite having a designated workspace, Heikki is moving across Finlicious's work areas more than the other cooks.
The Sous Chef is normally second in command but Finlicious's permanent Sous Chef was on a long-term sick leave during the data collection and got replaced by a leased employee, "Ryan". Ryan has been working as an "extra" chef/cook at various D&C Services locations (including Finlicious) for nearly a decade. Even though employee leasing has been criticised for lack of job security, Ryan prefers this type of employment due to its flexibility and changing work environments. Despite years' worth of work experience from Finnish kitchens, Ryan mainly uses his dominant language, English, at work. The lack of Finnish knowledge is relevant to how he enacts his professional role as a Sous Chef and turning to other staff for assistance.
Next to Heikki (and Pekka) we find "Ibou" and further away "Mali". Both of these employees are trainee cooks who are supposed to be mentored by Heikki. Ibou, however, is informally co-managing the migrant employees. We return to these employees in our excerpts and show how their different language capital is made relevant to self/other positioning in the group.
The 'cold food area' is where two Finnish colleagues, "Sini" and "Vilma", work. They are responsible for Finlicious's catering services for business clients. These two usually interact the most with another Finnish employee, "Nina", who is in charge of baked goods (in the middle). This interaction in Finnish is significant as we show in Excerpts 1 and 2.
As will be illustrated in the analysis, the non-L1 Finnish speakers (Appendix 2), Ryan, Mali, Akene, and partially Ibou, are positioned as outsiders in terms of participating in (1) social interaction with their L1 Finnish colleagues, and (2) the material spaces. We revisit and unpack this further in the upcoming sections. Those topographies of language practice are in effect topographies of inclusion and/or exclusion in a workplace. Ryan and Mali, appear to avoid contact with their Finnish colleagues, and instead turn to Ibou for help (Excerpt 1). In the process they follow a systematic, and emplaced, pattern that is enacted in each situated encounter. This is done in and through their position in the material space ( Figure 1) and language choices that perpetuate (or challenge) dominant language ideologies that circulate in their workplace. These employees have created together their own social space at the front-of-house where they speak and joke in English and Finnish ( Figure 5). Being outside the kitchen space means that the employees have more freedom as they are not monitored by the Executive and Head Chefs or the Restaurant Manager.
To sum up, Finnish undoubtedly holds a prestigious status in Finlicious. The excerpts below show how dominant languages are legitimised as such through spatial and social segregation (Lugosi et al. 2016) and stigma (Dong and Blommaert 2009) associated with lack of language proficiency. In addition to Finnish, English has a clear capital and employees with access to both languages are in a position to bypass strict hierarchies as per below. This succinctly captures the language ecosystem in Tasty Co with English as the 'official' language policy coexisting with the D&C Services 'unofficial' ones (see Angouri 2013 on strategic ambiguity).
In the following two excerpts we see Mali and Ibou negotiating their professional roles in and through their spatiolinguistic practice. 'Language' is both the means and a site for negotiating in situated encounters their relationship with others in their work environment.

Excerpt 1: Constructing an (un)supportive work environment
At the time of data collection Mali was undertaking an English-taught culinary degree in Helsinki and the main language of instruction during her internship at Finlicious should have been English. This was agreed by Heikki (her mentor) and her teacher from the culinary school. However, in practice our data shows that the host language, Finnish, is holding both prestige and the dominant role in interactions and decision-making power. Access to Finnish constructs and maintains social and physical boundaries between migrant and local employees. We show these (in/)visible borders through Mali.
Context: Mali notices that some ingredients are not stated on food labels. Labels are important artefacts due to diners' possible allergies and food restrictions. The following interaction starts when Ibou approaches Mali in her workstation, while she is busy filling in the labels: Mali shows initiative by starting an important work task. In the completion of the task, she is struggling with Finnish vocabulary. In her attempt to finish it, she draws on a resource available to her; Ibou. The excerpt is indicative of the close but asymmetrical relationship between the two employees (note the teacher-student role from line 6 onwards and in Excerpt 2), and shows the spatiolinguistic significance of the activity at hand. In detail (Figure 2), Mali's workstation is at the edge of the kitchen. Before the interaction above, when Mali noticed the blank labels, she took the labels and markers needed for completing the task and moved even further into the quieter top-left corner. This is illustrated in Figure 3.
Mali's movement can be read as evidence of her attempting to find a quieter, and possibly safer space. The corner is protected in terms of being farthest from her Finnish colleagues with whom there is tension as our next example will illustrate. In the far end of the kitchen, no one can hear her interactions with Ibou; she can ask for advice and make mistakes without judgment. When Ibou approaches Mali in her 'safe space' she uses body gestures to signal her intentions of writing the list of ingredients herself by pointing at the labels, asking for "a minute" (lines 2-3 and 5) and holding the marker. Mali had set herself a goal to learn one new Finnish word each day and her colleagues, including Ibou, knew about it. Here Ibou enables Mali's learning by instructing her letter by letter on (what he believes is 2 ) the correct spelling (lines 31-34), providing positive feedback (line 6) and responding with laughter to Mali's expressions (line 30). Mali, in turn, appears to enjoy this support which can be detected in her exclamation of joy and overt expression in line 29 ("wohoo!") signalling a sense of accomplishment after successfully writing seven Finnish words with Ibou's assistance (lines 7-28 and 35-52 omitted for space). They have successfully constructed team belonging which is frequently occurring in our data.
A supportive environment, however, is not always the norm. Note the following incident, drawn from the analysis of fieldwork notes: Mali is on her way from the fridge to her workstation ( Figure 4) and passes by three colleagues who are socialising whilst loading the serving cart. The grey triangle in the figure illustrates the so-called 'social zone' that Mali has to cross in order to get back to her workstation.
When Mali stops in front of her colleagues to move the serving cart out of her way, her co-workers ask her to read out loud a sticker on the cart saying: "edustuskäyttöön" ("for representation purposes"). This is a phonetically complex word and it takes Mali several attempts, alongside peer comments that come under the pretext of help but seems to add pressure, to read it out loud. After successfully pronouncing the word, her colleagues applaud and smile at her. This could have been perceived as a sign of support until the following comment in Finnish: "[…] she can read but it's another issue whether she has any clue what she's reading" ("kyllähän se osaa lukee, mut on eri asia onko sillä mitään hajua mitä lukee"), which results in the Finnish employees laughing at her expense (see also in Barrett 2006 for mock language and racial inequality in a Texan restaurant context). This happens in Mali's presence who smiles back and retreats to her designated zone, a safe, peripheral and excluded place. In the data collection follow-up, the Finnish employees were asked about the incident; their actions were translated as supporting Mali's wish to learn Finnish. Considering that her colleagues did not explain the meaning of the word, 'edustuskäyttöön', this claimed teaching appeared to embarrass Mali instead of serving her learning needs.
It is everyday incidents, such as those presented here, that disempower and exclude minority, typically non-L1, individuals. These excerpts are in line with gatekeeping studies showing how employees pay a heavy price for being perceived as 'different' and not one of 'us'. Limited knowledge of the dominant language further exacerbates the us/them divide. On this, we argue that a focus on topographies of practice provides us with a holistic analysis of how boundaries are indexed spatiolinguistically and enforced (and sometimes challenged) in situ.
Strategic use of language resources enables individuals to bypass formal hierarchies and claim more senior positions in the professional ecosystem. Even though in practice English is starting to stabilise its position as lingua franca among many kitchen workers in the Capital Region, Finnish still carries ideological and political significance as a working language. Our next example demonstrates how access to dominant language influences power relations between non-Finnish speaking employees and feeds into a policy/practice continuum. Our protagonist this time is Ibou. Ibou is a good example of the nuances related to in/exclusion processes. Instead of approaching them as a mutually exclusive binary, a topographical approach enables us to capture the dynamic relationship between what the participants do (agency) and what is done on them (constraints) in a situated encounter (Figure 1). We discuss this below. Ibou's centrally located workstations and the strategic mobilisation of language resources enables him to act as a linguistic gatekeeper (see also Humonen andAngouri 2022, 2023) and keeper of spatiolinguistic boundaries. This position comes with authority that goes beyond his official role and is informally legitimised by his superiors enabling him to instruct English speaking migrant employees. To some extent this is associated with the Restaurant Manager's, the Executive and Head Chefs (claimed) limited proficiency in English. Our next excerpt illustrates the connection between language resources and non-Finnish speaking employees' agency.
Context: Ibou is on his way to his workstation and decides to stop at Ryan and Mali's workstations: Although kitchens are regulated places, Ibou is consistently spanning boundaries (Wall forthcoming) and 'patrols' workstations ( Figure 1 on the situated  moment). This excerpt provides an example of Ibou's enactment of power over non-Finnish speaking employees which is a common pattern in the data. He appears to be comfortable with claiming a more senior position beyond his institutional role. The pattern is consistent across the dataset. In this example, it takes the form of providing advice to the more experienced and hierarchically senior Sous Chef, Ryan (lines 1-6), then approaching Mali and questioning the instructions she was given by the Head Chef, Heikki. More specifically, Ibou issues a directive of adding oil which is strengthened with deontic modals "have to" and direct address form "you" (line 12), and he further reiterates his point in lines 15 and 16. He continues questioning Mali despite her attempts at telling him twice that she is only following Heikki's advice on "what to do" (lines 9 and 13-14). Ibou's interference and message is delivered in an authoritative manner (see e.g., line 12). Direct forms come with a high face-threat cost and are rarely used in work interaction, outside contexts of emergency or high pressure; however, they are not uncommon in professional kitchen contexts where they typically reflect kitchen hierarchies. Here Ryan and Mali's uptake allows Ibou to enact a more senior position through his interactional strategy. Ibou's access to language resources that align with the company's formal and informal hegemonies, is directly relevant to the way his spatiolinguistic interactions with his peers are managed.
This however is not always a straightforward act and gets challenged in situ resulting in a sensitive equilibrium. We elaborate through the excerpt below.

Excerpt 3: Crossing the material (and social) boundaries
Context: Ibou starts tidying up the kitchen, transitions into Ryan's workstation and moves away his gastronorm pans. This is challenged by Ryan: Designated workstations carry symbolic significance in the kitchen contexts. Language encoding is a core layer of this topography. In excerpt 3, there appears to be some tension between Ibou and Ryan. Our analysis suggests that this is caused by Ibou constantly crossing material (and by extension social) boundaries, which goes against the typical kitchen practice and hierarchies.
The excerpt starts with Ryan challenging Ibou after noticing that Ibou had crossed the boundary; i.e., moved into his workspace and started moving Ryan's material objects. When Ibou fails to hear Ryan's initial question (lines 1 and 2), he repeats himself in exaggeratedly slow way with a raised tone (line 3), which signals Ryan's frustration with the intern cook. To further support this interpretation, Ryan does not accept or acknowledge Ibou's response in line 4 and interrupts him towards the end of his utterance. He then rebukes Ibou for taking actions without communicating with him first (line 6). As a response, Ibou becomes more defensive between lines 7 to 9; the reinforcing devices following by short pauses ("come (.) look I tell you" (.)) strengthens the rhetorical effect of the utterance and gives the impression of Ibou being responsible for "everything". Ibou finishes his turn with an implication that Ryan's work contribution is somewhat inadequate ("so maybe you can just do this"). So is a useful device for controlling interactional agendas (Angouri 2018) and Ibou makes good use of it here.
This results in Ryan acknowledging Ibou's perceptions ("yes okay") but he repeats his wish for better communication four times during this short interaction (line 5: "tell me what you're doing", line 6: "you don't tell me", line 10: "let me know", and in line 13: "just let me know"). Avoiding open disagreements is common in professional contexts and here the possible conflict is minimised by using mitigation strategy in which Ibou provides a reflection (line 11) and rationale for his actions: "I'm not good to stand up for myself so I always work on some things" (line 12), and Ryan changing his tone from confrontational to more compliant.
In this excerpt, in line with the patterns in the wider dataset, Ibou claims being responsible of doing "everything" and "always working". This projects Ibou's understanding of his perceived higher role at Finlicious, which is partially enhanced by his linguistic gatekeeping acts. This is further enacted in the kitchen and the front of house topographies as his workstations are centrally located. These material zones make him more visible to his colleagues, and particularly in the front of house, the prime space for socialising. We explore this in further detail below.

The front of house topography
As the kitchen map (Figure 2) shows, the open space and managers' monitoring affords limited opportunities for private talk. This has pushed socialisation into another work area; the front-of-house (FoH) and specifically at Ibou's second workstation (see Figure 5). 3 Ibou is the only cook who moves from the kitchen to the FoH for the serving hours where he is responsible of cooking simple items like pizzas and hamburgers. As illustrated in Figure 5, the FoH is fully accessible to customers apart from Ibou's counter in the top-left corner. This space provides a temporary "oasis of peace" ("rauhankeidas" in Finnish) as described by Nina.
The light grey arrows illustrate the typical customer flow during lunch hours. The dotted patterns represent kitchen employees' most visited zones of activities, i.e., the movement patterns from the kitchen to the hot and cold buffet counters, and Ibou's grill station. The waiting staff are scattered around the FoH, and their tasks usually involve keeping the surfaces tidy, serving the food, getting refills from the kitchen/dishwashing room, etc.
Ibou's FoH workstation, or the "temporary oasis of peace", is used as the most suitable place for socialising and resting as employees go to/leave from the kitchen, enabling all kitchen staff to exchange brief interactions with each other. It presents a particularly significant place for the non-Finnish speakers as here they can talk freely in English (or Finnish) without the pressure of being heard and monitored by their senior colleagues and Finnish co-workers. As in Figure 1, the fixity of this safe space and the fluidity of movement and language choice, provides with a dynamic framework for reading the practices here. Although this activity is emplaced in Ibou's workstation, the interactions do not always involve him directly. Being part of the audience, however, enables him to have access to the social talk of two different teams; the waiting staff and the kitchen workers. This increases his social capital and allows him to be a repository of local stories (e.g., 'gossip' is significant for team membership, see Angouri 2018;Holmes and Stubbe 2003) which he can mobilise to permeate in/visible boundaries. It also shows how some employees in the kitchen are not only placed on the 'us or the other' binary but on a staged 'one of us and one of the other' spectrum. 4 Ibou's counter is a significant node in our topography where informal memberships, Finnish, non-Finnish, mixed ability speakers, waiting and kitchen staff, are talked into existence.
Since employees are 'exposed' to customer contact at the FoH, D&C Services management justifies the Finnish language requirement for this exact reason; i.e., the perceived need to offer better customer service in Finnish for the host society. In practice, the non-Finnish speaking employees replying to customer enquiries in English was not an issue as most of the diners were non-Finnish speaking expatriates and/or professionals coming from various multinational enterprises who are used to speaking English. In fact, L1 Finnish waitresses were observed to direct English (and French) speaking customers to Ibou's workstation on several occasions. This succinctly illustrates that although migrants might get penalised (Excerpt 1) for not conforming to the local (language) ideology, the actual needs of the users of the space, at least in this specific context, are much closer to the migrants' linguistic capital.

Discussion and concluding remarks
We have shown the ways employees in our case company negotiate group membership and professional self spatiolinguistically. This process never takes place in a vacuum: "The spaces are always someone's space, and they are filled with norms, expectations, conceptions of what counts as proper and normal (indexical) language use and what does not" (Blommaert and Dong 2010: 368). Negotiating access to the spaces and social groups is subject to 'proficiency' which is spatiolinguistic. In order to be granted membership, employees need to index and be accepted as one of 'us' in at least some domains of activity. This comes with claiming and placing self within the group's spatiolinguistic boundaries. In our study, knowledge of Finnish has a dominant status and seems to be a 'condition' for participation in aspects of professional life. However, despite learning the language, newcomers are subject to the same penalty Gumperz showed in mid 70s. The linguistic capital of migrant employees is still devalued against the dominant ideologies (Kroskrity 2004) and formal and informal hierarchies that valorise and favour some languages against others. Ideologies structure social behaviour, and in its turn social behaviour reproduces structure and contributes to power imbalances between different speakers. This is an emplaced, linguistic, process as discussed in Figure 1 and illustrated in the data.
Going further, the material space is made relevant through body movement to claims of professional role. This is an integral part of employees' negotiating the topography of practice. The illustration below ( Figure 6) shows the zones of activity, in/visible social boundaries, through which the employees navigate their daily routines. It also shows the multiple centres and peripheries that are negotiated in situ in busy spaces such as in the professional kitchen. Workplaces are polycentric; the zones of activity are both flexible to allow for emergent practice as well as stable and rigid to maintain the status quo outside the context of any one individual encounter. Employees' negotiation of daily practice does not take place in a sociopolitical vacuum, power asymmetries impose constraints on their agency (Angouri and Piekkari 2018).
The darkest two circles in our graph represent the 'decision-making zone(s)' that are constructed and maintained by the kitchen managers, i.e., the Executive Chef, Head Chef and Restaurant Manager. The lighter grey areas demonstrates the 'social zones' (cf., discussion on FoH for non-Finnish speaking employees). These Topographies of language practice at work are areas of contact and social exchange which take place in Finnish language only. Looking at the Figure 6, we see how Finnish language use and areas of contact correspond symmetrically to the epicentre of power. The non-L1 speaking employees are at the periphery with the exception of Ibou's occasional legitimised presence in the power and socialisation zones. Yet, even though he is in a better position than his colleagues, our data show that he is still outside of the core Finnish speaking 'in-group'. The relationship between those zones legitimises practice which feeds into formal and informal policies.
The role of the individual is critical in the environment of any workplace. This is not a relationship of equals and participation is often heavily regulated by the 'gatekeepers' of a system. Consider Excerpt 1 and Mali's role in Finlicious. The interplay between pre-existing constraints and individual agency in the spatiomaterial context of the workplace is ongoing and constantly reinstated (Figure 1). The coconstituting relationship between language use in its spatiomateriality adds a significant dimension in the dynamics of teams interaction that needs further probing.
Building on our discussion, we have shown that looking at language use in and through topographies of practice is particularly useful for holistically unpacking how individuals negotiate spatiolinguistically full/partial membership in the various communities they interact with at work. This highlights the need for spatially situated study and redefinition of language use at work which affords access to more layers of context of linguistic, and particularly multilingual, practice.
We also argue that IS and the Gumperzian legacy is appropriate for a critical interactional analysis of language topographies and provide a still robust foundation for theoretical advancements and methodological innovation (Angouri in prep). Although current studies looking into spatiolinguistic matters often take different methodological approaches, Gumperz's emphasis on a "closer understanding of how linguistic signs interact with social knowledge in discourse" (Gumperz 1982: 29) is, according to our interpretation, at the very heart of what we are trying to achieve.
Current scholarship is concerned with understanding fluidity and complexity in the relationship between language and space. Undoubtedly these are central concepts and in need of further theorisation. At the same time, research needs to also account for the stabilities and the norms associated with dominant groups in social and professional settings. As much as our research shows the agency of the speaker, dominant ideologies and language hierarchies are visible and relevant to daily reality at work. As put by Bourdieu and Wacquant (1992: 146) language proficiency is "not a simple technical ability, but a statutory ability".
Workplaces are built on ideologies which come with expectations of performance and impose norms and ways of doing. Negotiation of norms always takes place in a web of power structures and hierarchies which are (more or less) visible to the participants (Angouri 2018;Duchêne et al. 2013). This is part of a dynamic set of relationships whereby social behaviour structures language ideologies, and language ideologies can be fed by power imbalances. Capturing and analysing this complexity presents us with theoretical and methodological challenges which lead to useful debates for the development of our discipline. Topographies of language practice constitute topographies of in/exclusion as we have shown here. In/exclusion is not a 'work' problem, it is directly related to the society we live in and the social order we all play a role in perpetuating or changing. In order to address the complexity of the phenomenon, research synergies and multidisciplinarity are necessary. Multimethod spatiolinguistic ethnographic research is certainly relevant to this agenda.
To conclude, language is a remaining bastion of discrimination at work; on the list of protected personal characteristics (in the UK: age, gender, ethnicity, religion, disability) language, broadly understood, is a glaring omission despite being constitutive of all others (on language and identity at work Angouri and Marra 2011). In this paper we made a case for a study of topographies of language practice as a conduit for capturing the process of negotiating access and, more broadly, integration in the workplace and, by extension, socioeconomic order. Current research is adding to our inventory for describing these complex topographies. Further work is needed on creating conditions for an active collaboration with policy makers and employers who are committed to providing conditions for empowering employees to fully participate in organisational life and beyond. We close the paper by passing the floor back to Mali and her account of team meetings, held in Finnish:

Appendices
Appendix 1: Transcription conventions ___ Underline indicate emphatic stress, e.g. "do it now" [ Left square indicate an interruption (.) Pause (( )) Researcher's notes, descriptions and clarifications (?) Researcher's best guess of an utterance < > Vocal descriptions, e.g. <laughs>, <shouts across the kitchen> ? Rising/questioning intonation … Section of transcript omitted All names are pseudonyms. experience at Finlicious only, not the overall work experience in a professional kitchen. b The order follows participants' self-reported language preferences at work. Apart from Ryan, our non-local participants do not use their dominant languages at work. The L/non-L distinction is meant to indicate the diversity of the repertoire rather than our assessment of evaluation of competence, we are marking specifically Finnish and English on impressionistic assessment because of their significance in the workplace. 'Fluent' indicates the level of language proficiency for work purposes, i.e. if the employees are able to complete their work tasks in that language. 'Fluent' here is not measured against native speakers language competences. The evaluation 'limited' in inverted commas indicates the, often, problematic use of the term; we refer to limited functionality as evidenced in our dataset which results in the interpersonal dynamics we discuss (e.g., Ryan frequently turning to Ibou for assistance while also negotiating hierarchy in their material space). More specific information cannot be disclosed to protect confidentiality. c These employees were not included in the study.