Digital media, friendships and migrants’ entangled and non-linear inclusion and exclusion

In this article, I shed light on the role of digital media and friendships in migrants’ inclusion and exclusion. Connecting the empirical findings of a qualitative, non-media centric, multi-method study to existing research in digital migration studies, digital media and friendship as well as the literature on inclusion and exclusion, I propose the concept of entangled inclusion and exclusion. This perspective underlines migrants’ continuous (re-) negotiations of their situatedness in the (trans-)local spaces of the places they come from, move to and through. While playing diverse roles in migrants’ development and maintenance of (trans-) local friendships, digital media contribute to non-linearities in migrants’ subjective feelings of being included or excluded, making the processes involved more complex while also serving as material and symbolic signifiers of entangled inclusion and exclusion.


Introduction
In this article, I debate the role that digital media and friendships play in migrants' inclusion and exclusion. By connecting an in-depth empirical study with the scholarship on migration and digital media or 'digital migration studies' (Leurs and Smets, 2018), digital media and friendship and (differential) inclusion and exclusion, I develop the concept of entangled inclusion and exclusion. This perspective reveals the complex, dynamic and non-linear processes of migrants' inclusion and exclusion and their related situatedness in diverse and entangled social, material and technological spaces. Bunnell et al. (2012: 490, 491) describe friendship as the 'interpersonal relationship between two or more people that is voluntarily entered into and may be similarly dissolved' and a 'means through which people across the world maintain intimate social relations both proximate and at a distance'. Friendships remain an under-studied form of relationships in the social sciences overall (Straughan et al., 2021) and in the field of digital migration studies especially (Sinanan and Gomes, 2020). In this article I ask: how do digital media influence migrants' friendship-making and their related inclusion and exclusion? The findings underline the importance of friendships to migrants' inclusion and exclusion and they contribute to uncovering 'the granular, ambivalent, contradictory' (Awad and Tossell, 2021: 623) role that digital media play in migrants' everyday lives and their social relationships.
The analysis is based on a qualitative multi-method study with highly skilled migrants staying on employment passes or holding permanent residency in Singapore. The study combined interviews with mobilemedia methods, cognitive maps and repertory-grid methods as elicitation strategies. The combination of methods pursued an open, explorative and 'non-media centric' (Smets, 2018) approach, which is 'imperative for digital migration studies to avoid repeating the sensationalist exceptionalism surrounding the technological fetishization' (Leurs and Smets, 2018: 8).
As I will come to argue, while playing diverse roles in migrants' development and maintenance of (trans-)local friendships, digital media contribute to non-linearities in migrants' subjective feelings of being included or excluded, making the processes involved more complex while also serving as material and symbolic signifiers of entangled inclusion and exclusion. It is important to consider migrants' own perceptions of being included and excluded and the involved processes of discursive difference-making in choosing and maintaining friendships. In any case, an analysis of the fostering of friendships and the role of digital media therein must be contextualised in regard to the complexities of migrants' situatedness in messy and contested everyday (trans-)local spaces.
Migration, friendships and the role of digital media Kathiravelu and Bunnell (2018: 491) highlight that in order to understand migrants' embeddedness in urban contexts, analyses of larger structural frames of migration legislation as well as categories and differences in hierarchisation must be amended by looking at 'other more dynamic, elastic, latent and surreptitious forms of affinity, relatedness and connection within the urban environment' such as friendships. Friendships were found to be crucial for emotionally, socially and culturally arriving and becoming embedded in a place (Bork-Hu¨ffer and Peth, 2020), as a network of support and care (Sinanan and Gomes, 2020) and for migrants' overall well-being (Bunnell et al., 2012;Straughan et al., 2021). Networks of friends can influence migrants' self-identification, resources, job situation, preferences and lifestyles (Sinanan and Gomes, 2020;Straughan et al., 2021). In contrast to family relations, friendships are deliberately chosen and need to be nurtured to be maintained, and can thus come with tensions and ambivalences (Hall, 2019;Kathiravelu and Bunnell, 2018;Sinanan and Gomes, 2020). Understanding migrants' friendship-making is an important aspect of 'capturing the diversity of migrants' relationships' (Sinanan and Gomes, 2020: 674) and, as I will show, of grasping their perceived feelings of being included and excluded.
The role that digital technologies play in migrants' connectivity and relationality has been a centre of concern in the digital migration scholarship (Diminescu, 2008;Leurs and Smets, 2018;Smets et al., 2020). However, major foci around concepts such as the 'connected migrant' (Diminescu, 2008), 'diasporas in the new media age' (Alonso, 2010) or digital diasporas (Brinkerhoff, 2010), have been the transnational maintenance of family relations as well as care relations (e.g. Madianou, 2012;Wilding and Baldassar, 2018). Friendships, particularly those in the receiving society, have only been the explicit focus in very few publications in this strand of literature (Sinanan and Gomes, 2020;Zhan and Zhou, 2020). Simultaneously, there is a broader scholarship that has looked into the role of digital media in friendship-making, but without explicitly studying migration contexts (see for example : Byron, 2021;Thulin et al., 2020). Still, there are various important insights in these existing studies in relation to this article's topic that will be elaborated on further.
Digital technologies influence, among others, the communication, information, job finding and economic situation, mobilities and organisation of everyday life of migrants (Alinejad and Ponzanesi, 2020;Diminescu, 2008;Kaufmann, 2020;Peth and Sakdapolrak, 2020;Zhan and Zhou, 2020). Hence, they can provide them with new autonomy, knowledge and/or access to services (Diminescu, 2008). The advent of the mobile phone was a particular game changer in many migratory contexts (Kaufmann, 2018). It contributed to the 'increasing prevalence of an ''always on'' culture of ubiquitous connectivity' (Madianou, 2016: 183). Digital media can thereby create '''ambient co-presence'', which is the peripheral yet intense awareness of distant others (Madianou, 2016: 183). Ponzanesi (2020: 1) explains that this has produced an 'affective turn online' in which digitally mediated spaces have become of central importance for the maintenance of social and emotional bonds elsewhere (Diminescu, 2008), for the extension of care practices (Byron, 2021) and for coping with acclimation in the host society (Hillyer, 2021). Migrants often use several media in parallel for different purposes and with different groups, while carefully weighing social and emotional consequences of the choice of each medium in the context of all other media -a phenomenon described as polymedia by Madianou and Miller (2012: 170). At the same time, the devices along with their migrant users are also subject to local representations and renegotiations (Platt et al., 2016).
Digital media have also been found to assist with the establishment of friendships in socio-material space upon arriving in a new place (Bork-Hu¨ffer and Peth, 2020), whereby Patterson and Leurs (2019) noted that digital communities can leave migrants with more choice regarding whether to join, compared to offline networks. Some authors argue that digital media can support the development of social contacts across difference, which expatriates were found to do more quickly than other migrant groups (McPhail and Fisher, 2015;Patterson and Leurs, 2019). Furthermore, research on digital media and friendship has found that mediated connections are characterised by strong homophily, where class boundaries are seldom crossed (Courtois and Verdegem, 2016;Thelwall, 2009;Utz and Jankowski, 2016). As will be further elaborated in the next section, digital media play an important role in the discursive (re-)production of categories of difference and ascriptions of identities, with central effects on migrants' differential inclusion and exclusion.

Entangled inclusion and exclusion and the role of digital media
The language used in relation to migrant inclusion, exclusion, integration and assimilation is politically loaded and influenced by discursive constructions and apparatuses that change in space and time (Schrover and Schinkel, 2013). Concepts such as integration, assimilation and immigration have long been criticised for connoting a permanence that migrancy often does not have, construing a homogeneous receiving society to which mobile subjects should adapt to varying degrees in a unitary and linear process (Gomes, 2018). They have been further deconstructed by perspectives underlining mobilities, translocalities, fluidities and temporariness in migration (Greiner and Sakdapolrak, 2013;Oswin and Yeoh, 2010;Peth et al., 2018). As a result of the criticism of normative concepts such as integration and assimilation, in migration scholarship the use of inclusion and exclusion as well as differential inclusion has become more common.
Underlining the frictions associated with inclusion using the example of the structural oppression and marginalisation of Filipino Americans in the US, Espiritu (2003) coined the term 'differential inclusion' to describe a form of forced incorporation, when populations were on the one hand defined as an integral part of a society but, on the other hand, only included if they accepted a designated subordinate position. Ye (2017) amended Espiritu's perspective on 'differential inclusion' by showing how, besides the role of top-down state migrant management, negotiations of social norms and civility as part of everyday encounters in shared spaces influence migrant inclusion. She concludes that inclusion and exclusion are not separated and distinct, but tightly interwoven, fluid and negotiated as part of everyday coexistence. I draw centrally upon this argument and argue that this often results in dynamic and non-linear processes of inclusion and exclusion of migrants.
In regard to the terms inclusion and/or exclusion, however, this article makes a point for using both, not only in an either/or sense but especially in regards to their entanglements as part of migrants' negotiations of their complex everyday spaces in the receiving societies. Given, among others, the arguments by Schrover and Schinkel (2013) that most migration legislation is designed with the aim of excluding, and Pe´coud's (2013: 2) point that nation-states' 'migration management' is mostly targeted at 'disciplining human mobility and the establishment of an ideal mobility regime in which control remains fundamental', using only 'inclusion' in a way minimises many migrants' experiences of the difficulties of arriving and living in the places they moved to or through.
Then again, the inclusion and exclusion debate often primarily tends to focus on externally steered processes and structures of migrant incorporation or expulsion and to a lesser extent looks into migrants' own agency, experiences of belonging and endeavours to be included or else to withdraw (Alinejad and Ponzanesi, 2020). This article makes a point for considering contradictory and entangled ways of inclusion and exclusion as part of migrants' situatedness in complex everyday (trans-)local spaces. Altogether this leads to dynamic, non-linear processes of inclusion and exclusion. Hence, while in this article a particular light is shed on the role of digital media and of friendships in regard to inclusion and exclusion, I strive to contextualise migrants' situatedness in complex everyday (trans-)local spaces in this process.
When looking into the role of digital technologies in processes of inclusion and exclusion, among others, digital geographers as much as communication scholars have underlined their important role in 'sociospatial processes of exclusion, adverse incorporation, impoverishment and enrichment' (Elwood, 2021: 209). These studies have shown how axes of inequality related to origin, race, gender, age and/or sexual orientation are reproduced through sociotechnological codes and algorithms (Elwood and Leszczynski, 2018;Valentine, 2006). Cover (2016: 112) pointed out that 'diversity online remains problematic. [...] the Internet and other digital sites of interactive engagement tend to be dominated by a more narrow set of representations'. The repetition of such representations by dominant groups is ultimately reflected in identity attributions, stereotypes, prejudices and the exclusion of non-dominant groups.
The digital migration scholarship has emphasised that digital media's affordances can come with 'uncomfortable' impositions (Awad and Tossell, 2021) and produce new challenges and costs in migrants' everyday lives. Digital platforms can be sites of rigid peer culture and pressure (Mainardi, 2020), make people feel entrapped (Awad and Tossell, 2021) and can extend social obligations and expectations into spaces of absence. Data extraction and biometric categorisations allow surveillance and monitoring of migrant bodies and result in their differential exclusion and harm (Madianou, 2021;Ponzanesi, 2020). Yet, Collins (2021: 869-870) underlined that socio-technological infrastructures as a 'middle space of migration' -which can also 'include heterogenous actors, institutions and technologies that may not necessarily be targeted towards migration itself' -can affect migration in diverse ways, possibly leading to its enabling, conditioning or control. Taken together, the summarised existing research suggests that digital technologies add to the complexities of migrants' inclusion and exclusion. The findings section exemplifies how this results in entanglements and non-linearities in migrants' inclusion and exclusion.

Migrant-led diversification versus migration management and the position of highly skilled migrants in Singapore
Given Singapore's history as a trading hub with an active trading community, migrantled diversification has been an integral component of the later city-state's path from its very early beginnings (Yap, 1999). Shortly after Singapore gained independence in 1965, the Singaporean government began to promote immigration in order to boost economic development. In international comparison, the approach to migration that Singapore developed became one of the strictest regimes of migration management. It establishes forward-looking control of the inflow of migrant subjects, and strongly manages the various migrant groups once they arrive in Singapore with a flexible and targeted regulatory policy (Gutting, 2016;Yeoh, 2004Yeoh, , 2006. Based on education level, professional field and income, migrants can apply for different temporary residence passes, while the rights and restrictions connected to each pass category vary substantially, with huge consequences for migrants' individual freedoms and livelihoods (see Bork-Hu¨ffer and Peth, 2020). Highly skilled migrants are eligible to apply for employment passes. They are separated from lower-qualified so-called 'foreign workers', who are issued work permits, as well as from skilled workers staying on Spasses (see Yap, 2015;Yeoh, 2006). Highly skilled migrants are privileged regarding the facilitation of their immigration procedures, access to local social services and their personal freedoms (e.g. having a choice of place of residence, the right to marry, whether to have children, etc.) when compared to the wide continuum of work-permit and S-pass holders (Baas, 2017).
Since the end of the 1980s, a succession of measures has been initiated to promote the migration of highly skilled migrants, referred to as 'foreign talents', to Singapore (Iwasaki, 2015). As they were seen as key to the country's pan-Asian and global economic competitiveness and the creation of new job opportunities, they were also encouraged to acquire citizenship or permanent residency (Beaverstock, 2011;Ho, 2006;Yeoh, 2006). Starting from the 2010s, and particularly after the publication of the 'Population White Paper' in January 2013, the public stance regarding immigration changed. In it, the government laid out plans to further promote immigration to accommodate economic and demographic demands (National Population and Talent Division, 2013: 6). This elicited growing resentment among the citizen population, especially expressed on social media platforms due to the restrictions placed on freedom of expression and the press (Bork-Hu¨ffer, 2017; Gomes, 2013;Lai, 2012;Yeoh and Lam, 2016). Responding to public resentment, the government announced a 'Singaporeans First' strategy, which led to immigration and residence conditions for highly skilled migrants in Singapore being tightened up (for a detailed account of the measures see Bork-Hu¨ffer, 2017).
As Zhan andZhou (2020: 1668) note, the result of this was a significant narrowing of pathways to permanent stay. Despite their privileges in comparison to work-and S-pass holders, highly skilled migrants also face limits on permanent stay and inclusion in Singapore. They have a broad range of incomes and contracts that may or may not include economic benefits such as housing or school allowances. High prices in the private housing market, limited access to public schools and high tuition fees in international schools are a burden for highly skilled migrants -particularly those with childrenwhose incomes are close to the boundary line for employment pass-holders and who do not receive allowances (Bork-Hu¨ffer, 2017). Furthermore, termination of contracts or unemployment leads to immediate cancellation of temporary passes. Such demands on the flexibility of labour create employment insecurity and settlement uncertainty as well as 'new conditions for exploitation and exclusion regardless of skill levels' (Zhan andZhou, 2020: 1657). Furthermore, Singapore's specific biopolitics of identity and belonging (Amin, 2012(Amin, , 2013Hall, 2013;Solomos, 2013) mean that highly skilled migrants too experience forms of institutional exclusion and social discrimination (Bork-Hu¨ffer, 2017;Yeoh and Huang, 2011). All these factors relate to their prospects for staying and to their practices of establishing friendships, as will be explained in the following sections.

Research design and methodology
The empirical data were collected as part of a larger research project, which examined the interplay of (im-)mobilities, digital media and the sense of place of migrants in Singapore. The study applied a qualitative multi-method design and included 50 transnational highly skilled migrants of different origins who migrated to Singapore, who stayed on an employment pass or who obtained a permanent residency. It combined elements of narrative-biographical and in-depth interviews with the smartphone interface method, cognitive maps and repertory-grid as elicitation strategies. The three additional methods helped deepen the conversation in regard to mobile media (non-)appropriation and patterns of use (smartphone interface method) and sociomaterial spaces of particular social and emotional importance (cognitive maps and repertory grid). The smartphone-interface method (Kaufmann, 2019) helps to prompt conversation about mobile-devices applications that migrants interacted with regularly, including those for communicating with and relating to others. Cognitive maps provided insight into and stirred conversation about the study participants' everyday spaces, including meeting spaces and places of emotional attachment. The repertory-grid method, following Tan and Hunter (2002), prompted participants to make spontaneous associations with places of importance. Overall, the triangulation of these methods pursued the goal of an open, explorative and non-media-centric approach to researching migrants' inclusion and exclusion.
The sampling took place using a combination of snowball and purposive sampling strategies. I made sure to include interviewees working in various occupations representative of highly skilled migrants in Singapore. Interviewees thus were employed by multinational companies (MNCs), local companies, government or educational institutions from Singapore or the countries in which they held citizenship, nongovernmental institutions (NGOs) or were self-employed. The dates of their respective migrations to Singapore ranged from several decades to just several weeks prior to the interviews. Interviews were conducted in German, English and Chinese and quotes used in this article were translated into English. Pseudonyms are used in the presentation of the results. I met some of them repeatedly to gain a deeper account of their migration processes, their media use and entanglements in everyday spaces as well as how they had changed, so that the total time I spent interviewing the participants ranged from 30 minutes to 6 hours. At the time of the study, all participants owned a computer, almost all had a smartphone and many had a tablet. The media they most frequently used for reaching out and for developing and maintaining friendships were WhatsApp, QQ, WeChat, Facebook, Skype, SMS and email. These offered different ways for communicating through text-based webpages, videos, chat and instant messaging, voice-over-Internet Protocol (IP) as well as text-based emailing.
A two-step process guided the analysis and interpretation. In the first step, Mayring's (2000Mayring's ( , 2005 'inductive category development' helped with identifying sequences within interviews with direct relevance for the focus on friendships out of the large data sets collected, and for identifying the breadth of differences in answers (e.g. in regard to the types of friendships and digital media used). In the second step, I employed narrative analysis (see MacKian, 2010;Wiles et al., 2005), which is suited to revealing the entanglements of subjectivities, trajectories and historicities (Wilson, 2016). Its aim is maintaining the 'many layers of meaning at which an interview operates' by connecting the 'intimate details of experiences, attitudes and reflections to the broader social and spatial relations of which they are a part' (Wiles et al., 2005: 90). In order to meet the demands of situating the role played by digital media and friendships in inclusion and exclusion in the complexities of migrants' everyday lives within the frame of this article, in the subsequent presentation of results, I exemplify my arguments by describing the trajectories of selected participants, while repeatedly comparing these to the overall sample.

Digital media, friendships and the entanglements and non-linearities in migrants' inclusion and exclusion
Overall, friendships played an important role for most of my study participants' subjective feelings of being included or excluded. The role that digital media played in this process was complex. The specific use of digital media for building and maintaining relationships that my study participants described was entangled in the complexities and temporalities of participants' socio-material spaces, encounters and friendships, their family situations, education and work contexts. It also related to their use of polymedia for various other purposes, for example work, navigation, transportation, entertainment and information. Their resulting feelings of being included or excluded were entangled and dynamically conditioned each other, which also contributed to nonlinearities in inclusion and exclusion.
Lifen (30 years old) had moved from Southeastern China to Singapore 2.5 years before our interview, had first finished a Master programme at the National University of Singapore before taking up a job at a local company. As for many of the participants in my study, her social experience of arriving in Singapore was complex and shifting. It ranged from first acquaintances and encounters that played an important role in her arrival (such as a roommate she had lived with for the first three months, landlords who supported her, initial circles of classmates -some of whom she started building friendships with and regularly met with until some left after finishing their studies) to being more focused on her new local colleagues after taking up her new job. A few months before I met her, she had given birth to her first child in Singapore. Starting with her pregnancy, and during her first few weeks as a new mother until I interviewed her, her offline social life was very much put on hold. When I talked to her, she strongly intended to obtain Singaporean citizenship once eligible, in order to gain a 'certain freedom' from China. However, she thought her chances of approval were small and considered moving to Hong Kong or back to Mainland China after staying for another 5 to 10 years in Singapore, as this gave her the advantage of being closer to her and her husband's parents, who also live in Southeastern China.
As for most of my participants, digital media were decisively entangled throughout Lifen's journey, although their usage patterns and their impact on inclusion and exclusion were non-linear and shifting. First, digital media served as a source of information before arriving (particularly through platforms such as Contact Singapore and ShiCheng BBS) and when navigating her first days in Singapore. Later, she used social media (especially WeChat, Facebook and WhatsApp) to connect with her first contacts and to make friends in Singapore. Meanwhile, social media also played an important role in maintaining connections with friends in China, which rather kept her from mingling with people in Singapore: Still, after she started work, she established relationships in Singapore and digital media became important for maintaining these contacts, for example through arranging offline meetings, and in the form of short exchanges that bridged the time between offline meetings. Her relationships with diverse others based in Singapore changed decisively after she gave birth. At the time of the interview, her life as a young mother mostly consisted of going to work during the day, and family life at home. At that point in time, digital media gave her the opportunity to stay in contact with friends remotely. Thereby, those friends she had been more intensively digitally connected to previously -many of whom were located abroad in China -became more important again, whereas she lost connection with those people she had mainly seen in offline space in Singapore prior to becoming a mother. It surfaced that the various digital connections she maintained (see also quote in the next section) provided a subjective feeling of being socially included -although the majority of these connections were with people based in China rather than Singapore. Lifen's example strongly emphasises the entanglements and the role of digital media in non-linear inclusion and exclusion. On the one hand, digital media helped her stay connected and included with some of her former friends in China. Although this helped her cope with loneliness during her early period in Singapore and also later when motherhood significantly changed her opportunities for socialising offline, in both phases it also kept her from mingling and becoming included in Singapore to a certain extent. On the other hand, digital media played a role in making friends and staying connected in Singapore. Moreover, this process of entangled inclusion and exclusion was deeply embedded in the specificities and complexities of her everyday spaces and was influenced by her previous relationships in her hometown in China, their translocal maintenance through social media, her intention to potentially return in the future, the relationality of her workplace, the constraints on maintaining socio-material friendships entailed by parenthood and her related return to mediated connections to friends in China.
Lifen's example furthermore suggests that the maintenance of friendships online can facilitate an emotional sense of inclusion for migrants that is not (only) bound to the (temporary) migration destination. Hence, Lifen's case underlines how feeling included or excluded has a deeply subjective dimension. Although there were other research subjects, who said they benefitted from the emotionally important connections elsewhere, other participants struggled with social pressures to stay in frequent online contact with friends and family outside of Singapore, at least temporarily. They felt deprived of possibilities to make contacts locally and of being included in Singapore as a consequence.

Digital media and differential inclusion and exclusion in migrants' friendship-making
My empirical data reveal how the process of arriving in the migration destination and of establishing and maintaining friendships is entangled with processes of difference-making. It highlights how migrants' specific choices of friends contribute to their own differential inclusion and exclusion as well as their practices of differentially including and excluding others. I will particularly draw on Malte's as well as Lifen's account to substantiate these arguments.
Malte (33 years old) is an engineer, whom I met twice over a period that extended across his first year and a half in Singapore after he moved there from Southern Germany. He held a flexible contract that allowed him to return any time to his company's base in Germany. On our first meeting, he was struggling with the financial constraints that the expensive city had on his living arrangements, as he found himself having to share an apartment in order to afford a place to stay, which was something he was not willing to continue to do in the long run. On our second meeting, he described his strategies for finding friends since arriving:  [...] And there was, yes, I would say that people mainly ... Well, I thought it was important to get to know people mainly through sports, because I just wanted to make sure that I could somehow connect with people, for example to play soccer. Besides that, of course, you also say, well, you go out for a drink or something in the evening or you meet up to watch soccer or something like that. Well, these were mainly Germans, more from Facebook, I would say.
The interview excerpt shows how Malte categorises his contacts through a mix of constructions and ascriptions of origin and ethnicity. The social groups to which he belongs in social media are partially divided according to these constructions -or at least he makes this distinction: 'Germans, more from Facebook' or 'mixtures of Singaporeans and [...] Asians who moved here' from WhatsApp. Origin, citizenship and ethnicity were pervasively stressed as identifying labels of friends by all interviewees. It must be noted here that Singapore is an example of particularly strong state-discursive productions of ethnic containers through its CMIO-categorisations (Chinese-Malay-Indian-Others) of persons with Singaporean citizenship (Chua, 2003;Ye, 2016). Ye (2017Ye ( : 1033 emphasises how this contributes to 'the normative and productive categorisations' of race and citizenship in the city-state. It has been suggested that the categorisations likely promote subtle discursive (re-)productions of demarcations between citizens and migrants as well as between migrants from different origins (Bork-Hu¨ffer and Yeoh, 2017;Gomes, 2013). By choosing who to reach out to through specific platforms based on difference, Malte himself draws boundaries as well as (re-)produces differential inclusions and exclusions of others.
Malte's example shows how online relationships are entangled with offline relationships (e.g. playing sport, going out for a drink and meeting to watch sports) and how both reinforce each other. For Malte, employing digital media was highly important for establishing connections and friendships with diverse others in online and offline spaces and his subjective feeling of being embedded and included in Singapore. In contrast to Lifen, Malte explicitly linked his feeling of being included with having succeeded in establishing friendships with Singaporeans as well as with other migrants.
I draw once more on Lifen's account to exemplify how the differential choice not only of friends but also of the specific polymedia used to connect with those friends results in the differential inclusion and exclusion of others. Lifen divided her contacts into 'Chinese friends' and 'non-Chinese friends' and noted that she engages with them through a structured set of polymedia: T: Since you are in Singapore, do you also use social media? Lifen expressed a very clear approach as to which medium she uses to connect with which group of individual friends -some based in Singapore, some abroad -as part of her set of polymedia. Furthermore, she reports using each medium with a different frequency. Whereas she checks Facebook and WhatsApp, through which she connects with her 'non-Chinese friends', on average only once a day, she uses QQ, through which she connects to her 'Chinese friends' in China, more often. Again, she connects with her 'Chinese friends' in Singapore via WeChat on average only once a day. This eventually affects the types, qualities and depth of relationships she has with these groups of friends. It is also noteworthy how the actual location of friends (Singapore or abroad) matters less in her account than their ascription as Chinese or otherwise.
Overall, the friendships that my participants often established, but not always, included subjects of various origins. Those interviewees who were not embedded in transnational networks prior to migrating were more likely to stay within entangled networks with others of similar origin. In their attempt to establish new contacts after their migration to Singapore, strikingly, all research participants established lasting contacts with other highly skilled subjects, so there was hardly any cross-class sociability. Reaching out to others through platforms they had used before migrating contributed to homophily.

Digital media as signifier of inclusion and exclusion
In my data, digital media not only played a role as subtle agents for enabling and structuring social relationships as described above, they could be the central focus, that is signifiers, of negotiations of inclusion and exclusion, thereby contributing to their entanglements, as I will show by drawing on different examples. Christine, a 28-year-old migrant from Germany, had resisted buying a smartphone prior to coming to Singapore out of distrust regarding data security. However, a few weeks after arriving in Singapore, she feared being excluded if she did not buy a smartphone and started using particularly WhatsApp as a way to keep in contact with her new peers. However, the possibility of using digital media to build connections was welcomed by most of my study participants rather than perceived as social pressure. For some participants they served as a natural, mundane part of blending in, of becoming socially included.
Digital devices and media could also act as material as well as symbolic signifiers of entangled inclusion and exclusion. Gerhardt (63 years old), who had been living in Singapore for about three decades at the time of the interview, explained how: My friends, when we meet and sit together, would have one or two mobiles lying on the table, and then, one of them would say: ''Oh, by the way, there is this and that app and that is an app that one should really have a look at''.
Here the smartphones, as material artefacts 'lying on the table', prompt conversations about specific applications that become signifiers of entangled inclusion and exclusion: being inclusive of those among the peer group who know and use them, and simultaneously signifiers of exclusion of those who do not have smartphones in the first place or who do not try out, know and/or use specific applications. This again underlines digital media's role in the entanglements of inclusion and exclusion.
Then again, smartphones could act as material signifiers of perceived cultural difference and differential inclusion and exclusion, as Andres' (43 years old) statement reflects: [people in Singapore] are not too eager to have a social integration with other colours, meaning they are more attached to their phones than other people so these are . I think the human interaction is missing. For me that's stressful because I came from a place where people talk every day, even in a small place we had to be with the people. For Andres, who had migrated from the Philippines to Singapore eight years prior to the interview to take a job in the non-profit sector, the smartphone, as a material artefact of digitally-mediated relationality, is seen as a marker of different cultural approaches to relating to others, and is demarcated from forms of offline socio-material sociability that Andres feels acquainted with and perceives as desirable. His example also highlights the subjective side of, and challenges involved in, friendship-making, when he notes: .if you stay here and you are with the community, social integration might be there but it's too stressful in a sense that if you don't really find courage to talk and to open up a discussion so there will be no communication.

Discussion and conclusions
In this article, I analyse the role that digital media and friendships play in migrants' subjective feelings of inclusion and exclusion. For the subjective feelings of becoming included as well as having 'socially arrived' in Singapore, the successful establishment of friendships was decisive for my study participants. While the importance of friendships and the role of digital media in supporting the complex process of emotionally and socially arriving in a place that migrants move to and through have been found to be of relevance in other studies, this article contributes three additional insights.
First, as part of migrants' situatedness in complex and contested (trans-)local everyday spaces, their inclusion and exclusion are constantly (re-)negotiated, which leads to entanglements and non-linearities of inclusion and exclusion. As I have shown with the case of Lifen, digital media contribute to non-linearities in this process. Her shifting practices of reaching out to translocal contacts outside of Singapore and strategies for building friendships in Singapore through digital media resulted in a non-linear process of inclusion and exclusion in Singapore. Furthermore, her story exemplified that migrants' subjective emotional sense of inclusion is not necessarily (only) bound to the (temporary) migration destination but composed out of their overall translocal social situatedness. Digital media's potential to steer intense emotional-affective awareness of co-present others (Madianou, 2016) is central here. This finding once more challenges integration-focused perspectives that underline the necessity of migrant-local relationships for migrant incorporation and that focus only on inclusion in the migration destination. As Ponzanesi (2020: 8) argues, the current discourse must still more fully take account of 'the rise of the digital and digitalization, which are profoundly changing the importance of the nation-state for people in terms of belonging, zones of association, and everyday contact'.
Although digital media's subtle and banal structuring of everyday encounters and relationships has become an intrinsic element of migration, it is only one factor influencing migrants' entangled and non-linear inclusion and exclusion as part of their situatedness in messy everyday spaces. Other factors that together contributed to dynamic and nonlinear processes of inclusion and exclusion were state-led migration management and the prospects for staying in Singapore and in other potential destinations, variegated (trans-)local experiences with institutional inclusion and exclusion, migrants' own intentions to stay or move on, expectations of families and relatives and everyday social encounters and experiences of discrimination. Highly skilled migrants have been more the focus of state-led endeavours to increase inclusion in Singapore than migrant workers and S-pass holders. Yet, their temporary status in particular, and also for some the financial constraints on sustaining a family, mean that most do remain excluded from settlement (see also Zhan and Zhou, 2020), with implications for their endeavours to make friends locally.
Second, my research shows that it is crucial to understand the role of polymedia in migrants' experiences of, and also their own practices of, differential inclusion and exclusion in the process of friendship-making. While digital media, through their ubiquitous availability and accessibility to highly skilled migrants, on the one hand offered a way of bridging constructed boundaries and categories of difference, on the other hand those same media spaces were often still inherently structured according to just such boundaries and categories. Migrants were simultaneously engaged in a variety of online platforms, through which they connected with a more diverse set of friends. However, the -at least partial -structuring of digital spaces according to origin has an effect on the forms and qualities of relationships: each digital medium comes with different affordances and possibilities for relating to others. This can affect the qualities and depth of digitally-mediated relationships that subjects have with their contacts based on origin. The strong alignment of friendships according to class boundaries that my research found is in line with other studies on friendship and homophily (McPhail and Fisher, 2015). In the case of Singapore, migrants' boundarydrawing according to constructed categories of origin, ethnicity and class thereby strongly resonates with state-steered discourses of difference-making in the city-state.
Third, beyond their subtle and mundane entanglements with migrants' social relationships and everyday lives, digital media, particularly smartphones and specific applications, serve as more overt material and symbolic signifiers of (differential) inclusion and exclusion. These findings once more exemplified the tight entanglements of inclusion and exclusion, when such media and related artefacts were employed as signifiers of inclusion among those who know and use them and markers of exclusion of those who do not have smartphones in the first place or who do not try out, know and/ or use specific applications. Digital devices or specific applications could be interpreted as markers of different cultural approaches to relating to others, and discursively separated from other types of sociability that migrant subjects were acquainted with and perceived as desirable.
and Simon Bunchuay-Peth and the anonymous reviewers for their valuable feedback.

Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was financially supported by the Alexander-von-Humboldt-Foundation and the Asia Research Institute, National University of Singapore.