2542-4920 Glocalised Motherhood: Sociality and Affect in Migrant Mothers’ Online

2019 ABSTRACT This article explores the importance of relationships between migrant mothers in Australia through the lens of Facebook groups created by and for migrant mothers, referred to here as glocalised maternal communities. The combination of migration and motherhood creates a dual rupture in mothers’ social networks, leading to isolation and emotional challenges. Drawing on thematic analysis of interviews conducted with 41 mothers from a range of migrant communities (including Indian, Malaysian, Swedish, German, Brazilian and British), living in Melbourne and Sydney, I investigate why migrant mothers have responded to these challenges by creating and participating in locally-based, social media such as Facebook for mothers with a shared national or cultural identity. These communities provide resources and support for women’s migrant maternal projects, for example relational settlement and cultural transmission, and also meet women’s own needs for intimacy, friendship, and support. From an analysis of sociality, emotion, identity and relationality in these glocalised maternal communities, I develop a framework of multilayered and multimodal migrant maternal sociality, including a re-formulation of Haythornthwaite’s concept of ‘latent ties’ (2005) to highlight the affective benefits of online


INTRODUCTION
New motherhood can be a lonely time, and new migrants often struggle to make friends. The combination of migration and motherhood creates multiple ruptures in migrant mothers' social networks, often leading to social isolation and the construction of new forms of belonging and community (Gilmartin and Migge, 2016;Manohar, 2013aManohar, , 2013b. This article explores how migrant mothers use online communities to build relationships and resources for themselves and each other. I argue that migrant motherhood presents specific challenges relating to emotions, identity and sociality, and that migrant mothers in Australia have responded by creating and participating in 'glocalised' online communities for migrant mothers. 1 They are 'glocalised' because their salience is rooted both in their new locality and a transnational maternal identity that links that new locality with a point of national, cultural or linguistic origin. These glocalised migrant maternal communities provide resources and support for women's migrant maternal projects, for example relational settlement and cultural transmission. These migrant maternal projects are forms of 'motherwork' (Collins, 1994) formulated in response to their migrant context. Maternal norms and practices are contextually specific and dynamic (Collins, 2000;Glenn et al., 1994) and may therefore change due to migration (Gedalof, 2009;Ho, 2006;Liamputtong, 2006;Manohar, 2013a;Manohar and Busse-Caŕdenas, 2011). Migration may disrupt and reformulate maternal care practices (Erel, 2002;Hondagneu-Sotelo and Avila, 1997), facilitate maternal care (Ackers, 2004) or constitute a maternal care practice in itself (Westcott and Robertson, 2017).
To explore how and why migrant mothers use these online communities, this article draws on a thematic analysis of interviews conducted with 41 mothers from a range of migrant communities (including Indian, Malaysian, Swedish, German, Brazilian and British), living in Melbourne and Sydney, who create and participate in © 2019 by Author/s 3 / 15 their new maternal role (Bartholomew et al., 2012). Even in these digitally connected times, and despite the wellestablished concept of the 'connected migrant' who maintains ties with their social networks at home while being geographically mobile (Diminescu, 2008;Leurs and Ponzanesi, 2018), migration continues to disrupt migrants' social infrastructures. Studies focusing on the challenges of migrant motherhood often call for increased social support as a means to address them, but rarely analyse how that social support might be constructed (Benza and Liamputtong, 2014). This article applies an emotional lens to migration and motherhood, drawing on recent scholarship focusing on migration as an emotional experience, which is a useful counterbalance to the 'dominance of economic and political analyses of migration' (Boccagni and Baldassar 2015). Research in this field tends to draw on scholarship that views emotions as embodied, social, and relational (Ahmed, 2004;Burkitt, 2014), a framework also applied in this article. This article looks at how emotions operate in a multimodal maternal community, exploring how mothers draw emotional sustenance from both online and in-person interactions facilitated by the groups. In enabling multilayered and multimodal social interactions and affective exchanges, the groups act as social and emotional safety nets.
Concepts of identity are important for explaining why migrant mothers seek support in these particular groups. Identity is viewed as relational and intersectional (Collins and Bilge, 2016), formed and enacted in relation to 'imagined communities' (Anderson, 1983(Anderson, /2006. The women recognise each other as people who share experiences of migration, motherhood and childhood, share narratives of what it means to be a mother with a particular national or cultural identity living away from 'home', and share maternal projects rooted in those narratives and experiences. As these descriptions suggest, relationality is a key lens through which migrant mothers' experiences are understood; their identities, experiences, emotions and plans are deeply affected by their connections with their family, national and ethnic communities, local communities and other mothers. These relations are central to the way migrant mothers understand themselves and their experiences. Women draw on these relationships and shared narratives to create the support mechanisms they need to navigate the challenges of migrant motherhood. Furthermore, in centring their identities as migrant mothers, they can position themselves within an imagined migrant maternal community, of which the online group is a manifestation. This article draws on Kanno and Norton's reworking of Benedict Anderson's concept of 'imagined communities' (Anderson, 1983(Anderson, /2006Kanno and Norton, 2003). While Anderson used the term to describe a sense of belonging in nationstates that transcends tangible and immediate personal connections, Kanno and Norton link the concept of imagined communities to identity construction, suggesting that an individual's investment in an imagined community can drive choices and (re)frame experiences (2003: 248).
These concepts of sociality, affect, identity and relationality are brought together in the concept of 'glocalised motherhood'. Drawing on shared nodes of identity -motherhood, origin, current locality, language, culture -they manifest the relationships between each other into glocalised maternal communities, which are given a visible identity as online groups. Driven by the specific emotional challenges of migrant motherhood, these groups facilitate 'horizontal caring' (Francisco-Menchavez, 2018) between mothers and support migrant women's 'maternal goals' (Keller, 2010). Through an analysis of the ways in which migrant mothers make use of online affordances to obtain emotional sustenance online and facilitate in-person social relations, this article will outline a tripartite framework of migrant maternal sociality. The framework encompasses online and offline interactions and clarifies the significance of different modes of sociality produced by interactions in and around these glocalised maternal communities, including casual sociality, heartfelt friendship and 'latent ties' (Haythornthwaite, 2002).
This article follows women's social relations as they move between online and offline interactions, and it is important to note how integral online technologies have become to many people's experiences of motherhood (Arnold and Martin, 2016;S. Johnson, 2015;Lupton et al., 2016) and migration (Diminescu, 2008;Francisco-Menchavez, 2018;Komito, 2011). Studies of mothers' online practices suggest that a desire for knowledge and support from other mothers is a key driver of mothers' use of online maternal networks (Madge and O'Connor, 2006;Price et al., 2018). Although much has been made of mothers' fear of judgement from other mothers (Moore and Abetz, 2016), studies of mothers' online and offline information-seeking suggest that mothers prefer to seek information from other mothers with whom they have an 'experiential overlap' (Davis, 2015: 168). Experiential maternal knowledge is seen as trustworthy and relevant (Davis, 2015;Loudon et al., 2016). Scholars disagree about whether online communication is 'conducive to sociality' (Chambers, 2013: 16) or offers 'the illusion of companionship without the demands of friendship' (Turkle, 2011: 1). Feminist scholars of motherhood online tend to agree that online networks offer the potential for challenging normative ideals of good motherhood, but as products of those same socially-constructed norms, they also reflect, reinforce and intensify normative gender relations and norms of motherhood (Mackenzie, 2017;Madge and O'Connor, 2006).

RESEARCHING MIGRANT MOTHERS ONLINE
This research on migrant mothers and online communities presents a feminist, qualitative, sociological study involving semi-structured interviews with 41 migrant mothers in Sydney and Melbourne, Australia. Women from ten countries 4 across three continents were interviewed between February 2016 and January 2017. All were members of Facebook groups for migrant mothers in Australia, such as groups for Indian mothers living in Sydney, or Brazilian mothers in Melbourne. All Facebook groups in the study were based around a defined migrant identity, either national (e.g. British), regional (e.g. Scandinavian), or linguistic (e.g. Spanish-speaking). Although the study approached the groups as online phenomena, they are examples of an 'embedded, embodied and everyday' internet (Hine, 2015). All groups facilitated offline encounters as well as online interactions, although not all participants availed themselves of these opportunities. Participants were aged between 30 and 49, mostly married (to men), and had been in Australia between six months and 28 years. Of the 41 participants, 27 had become mothers for the first time in Australia; the others had arrived as migrants with at least one child.
All participants had stable and documented migrant status. Most held Australian citizenship or permanent residency. The few who were on temporary skilled working visas 5 expected to become eligible for permanent residence in the near future. The participants were drawn from populations with diverse relationships to Australia's migratory history. For example, while migrants from Britain and Ireland have had a significant presence in Australia since its colonisation, and remain the largest group of overseas-born people, migrants from India and Malaysia arrived in large numbers only after the end of the White Australia Policy in the 1970s (Jupp, 2007). 6 The Indianborn population has almost tripled since 2006, now numbering around 455,388 (Australian Bureau of Statistics, 2017a). Brazil represents a relatively small source of migrants to Australia, but sits alongside India, Pakistan and Nepal as one of the fastest-growing migrant populations in Australia (Australian Bureau of Statistics, 2017b). In 2017, Australian Census figures made headline news, with media outlets reporting that "nearly half" of people in Australia are born overseas or have at least one parent born abroad, and that the overseas-born population is now predominantly Asian-born rather than European-born (Australian Bureau of Statistics, 2016;Hunt, 2017).
While the Facebook groups are open to mothers of all social class backgrounds, the research participants were all broadly middle-class. In line with recent migration scholarship, I define the participants as 'middling migrants' (Luthra and Platt, 2016;Robertson and Runganaikaloo, 2014), that is, their migration was driven neither by economic necessity or escaping harm, nor by elite global careers. A third cited their relationship with their partner as a primary motivation, 7 and a further third cited their partner's work opportunities as their main reason for moving. Other common motivations included lifestyle, desire for travel and new experiences. Three women cited work opportunities for both themselves and their partner, and two had moved mainly for their own work. Broadening the definition of 'middling migrant' beyond the dominant labour market and migration motivation frameworks, the research also took into account factors such as their ease of access to mobile and internet-enabled computing devices, their ability to travel and make extended visits home, orientation to paid domestic work 8 , and access to transnational knowledge networks. 9 10 Taking migrant mothers' Facebook groups as a starting point for this research facilitates an emphasis on connections between mothers. It also establishes a temporally open approach to migration and motherhood: because the groups are used by mothers of both infants and teenagers, and by recently arrived and longer-settled migrants, drawing participants from the groups broadens the scope of the study beyond the common research focus on migrant women's experiences of pregnancy, birth and early motherhood (Hennegan et al., 2014;Joseph et al., 2018;Pangas et al., 2019). While making for a complex sample, this diversity brings together the experiences of women often analysed separately, or not at all, offering the opportunity for new insights, connections and contrasts. 5 / 15 In this research, all participants, including me as researcher, shared experiences of migration and motherhood. Most participants were, as I am, middle-class, tertiary-educated, married to men, and had migrated to Australia for multifactorial reasons including work, study, love and lifestyle. We were all 'fluent in Facebook' and English, with easy access to digital technology. In other areas, however, there was less common ground. Ethnicity, nationality, length of time since migration, children's ages, mother tongue, and current place of residence, were among the points of difference which affected our experiences of, and perspectives on, the research topics. This meant that a straightforward understanding of myself as either 'insider' or 'outsider' was impossible to maintain (Dwyer and Buckle, 2009). Shared social locations may be consciously deployed by researchers to negotiate access or build rapport (Duncombe and Jessop, 2012) and this deployment may be challenged by participants or 'strategically rescinded' by the researcher during the research process (Manohar, 2013c).
Interviews were chosen over content analysis or participant observation, both commonly used online research. This decision was both ethically and methodologically motivated. Drawing on Nissenbaum's concept of 'contextual integrity', which states that decisions relating to privacy online should take into account 'entrenched informational norms' specific to a particular context and 'general ethical and political principles as well as contextspecific purposes and values ' (2011: 38), I determined that the groups' norms and structures indicated a reasonable expectation that members' posted content would be observed only by group members. Preliminary research also suggested the groups might be used by isolated and vulnerable people, and that a researcher's presence may inhibit support-seeking or be negatively perceived by members (A. Johnson et al., 2018). Moreover, talking to migrant mothers about their online and offline experiences, and the meanings they ascribe to them, was central to the purpose of this research. Interviews revealed insights into the participants' online behaviour that could not have been gained through observation or content analysis. Focusing only on visible content privileges the perspective of the writer, over that of those who read it (Sun et al., 2014;Yang et al., 2017). Interviews have limitations too. Memories do not always tally with the evidence of the site and participants may impute motivations to other members which may not be reliable. Nevertheless, qualitative interviews focusing on women's experiential interpretations aligned well with the project's research objectives.

MIGRANT MATERNAL ISOLATION, HOMESICKNESS, GUILT AND CULTURAL TRANSMISSION
To explain why migrant mothers choose to seek support in glocalised maternal communities, this analysis draws together extant literature and new data to discuss the specific emotional challenges faced by migrant mothers. Experiences of isolation, homesickness, guilt and the emotions associated with attempting to pass on elements of culture and language to their children are closely interconnected. The close association of motherhood with loneliness and isolation is well established (Lee et al., 2017;Rogan et al., 1997). Investigations of maternal mental health recommend increasing mothers' social support to reduce isolation, increase practical support and thereby improve mental health (Hetherington et al., 2018;Leahy-Warren et al., 2012). Studies suggest that when migration intersects with motherhood, isolation and loneliness may be exacerbated. A 2014 meta-synthesis of fifteen qualitative studies relating to migrant mothers' experiences found that 'feelings of isolation, loneliness and depression from lack of support' were common (Benza and Liamputtong, 2014: 582). A 2011 literature review of studies on postnatal depression (PND) in migrant mothers in developed country settings found 'immigrant women are at particular risk of PND in their host country compared with women in the general population' with lack of social support cited as a key risk factor (Collins et al., 2011). The mechanism chosen by women in this study to fill this gap in social support -the creation of and participation in specifically migrant and maternal groups -highlights the salience of this particular combination of experiences and identities.
Online mothering communities have the potential to combat maternal isolation, and provide semi-public spaces in which women can interact even as they remain physically located in the private space of the home. Participating in locally-based groups increases the likelihood of frequent and repeated interactions with the same people, rather than relying only on physical encounters in local streets, shops, and playgrounds. For new migrants with few local attachments, those connections can provide a pathway out of isolation. Unlike many face-to-face mothers' groups, in which mothers meet during the day, online groups do not exclude mothers in full-time employment as they are accessible outside work hours, or during work hours using a smartphone, for example.
For women in this study, experiences of isolation were produced by differing intersections of elements including motherhood, migration and work. Women who migrated for their partner's work often found themselves experiencing an extended period as a 'stay at home mother' in a new country. Simran and Sunita, 11 both recent migrants from India, had left professional careers to have their first baby; soon after, they moved to Sydney for their partner's work. In India, as middle-class professionals, they would have expected to return to work after a 6 / 15 © 2019 by Author/s few months, supported by family and paid domestic workers. Simran found the transition to sole responsibility for all domestic tasks and childcare exhausting. Before she found the Indian mothers' group, she felt lonely, spending all day alone with her toddler. An Indian mother she encountered in the local shopping centre told her about the Facebook group and also a local WhatsApp group for Indian mothers in her area. Simran joined both groups, using them to chat and find answers to questions. She has met up with mothers from the groups, with and without their children. What she likes most about the groups is "feeling connected to people, lot of information flow (…) you feel like you're not alone, there are more people who are facing the same challenges as you are…". Simran and Sunita both keenly anticipated a time when they would be able to return to work. For the mothers, the simultaneous loss of working identities and networks centred around work, place, friends and family, created a strong sense of isolation, of having been 'cut adrift from the social moorings secured by affective ties of family and friends, as well as community and place' (Yeoh and Khoo, 1998: 172). Lisa, a white British migrant, arrived in Sydney pregnant with her first child. Her experience encapsulates the double-edged effect of motherhood as both disruptive and connective: With regards to an identity when I was coming over, I didn't feel as though I was the work person I was when I left, and then I was also pregnant, and I didn't really know how to handle that because it was the first time I'd been pregnant. And then I didn't have any friends here, so I didn't know how to handle that either! So it was weird. The first few months were really quite quiet, lonely, being pregnant, not wanting to do anything. (…) It wasn't really until I started my antenatal classes and going to meetings regarding the pregnancy that I started meeting people. So then I became that mum figure. I kind of fit into something then.
For Lisa, the combination of migration and emergent motherhood entailed a loss of her worker identity, and the removal of the workplace as a potential springboard for new social relationships. As the pregnancy progressed, motherhood provided her with a new source of social relations, focused around her new maternal identity. A secondary rupture in her maternal networks occurred when many of her new friends returned to work, making them unavailable as daytime companions. Lisa was unable to return to work, due to the cost of childcare 12 and lack of family support. Seeking new friends, Lisa joined the UK and Irish mothers' Facebook group.
Where women have become mothers after migration, motherhood has often disrupted the social networks they had begun, with difficulty, to establish. Women found their post-migration employment-centred social networks insufficient once they stepped outside the workplace as new mothers. Without a dense social infrastructure to fall back on, women found they knew nobody who could share their new concerns or the new child-centred rhythms of daily life. Young children's sleeping and eating routines limited women's ability to leave their home at certain times, meaning they often missed social events. As Tanja recalled, "you have a baby and all of a sudden you're not that mobile any more, or you have to stick to sleeping times." Some women found that differences in parenting styles, layered on top of broader differences in conversational and social norms, inhibited their ability or desire to make or maintain friendships with Australian mothers. Stefanie described "falling out" with her best friend, who was Australian, because of the differences in their parenting practices. In the German mothers' online group, she found women whose child-raising approaches chimed with hers, and had recently come to rely almost entirely on her circle of fellow German mothers for support on child-related issues.
Some women contrasted their isolation with the social support they imagined they would have had in their home country, particularly those who had witnessed more communal modes of motherhood, for example in India and South America. Even women from countries with similarly individualised and intensive modes of mothering to those encountered in Australia, imagined an alternative context with childcare and friendships on tap. For example, Lisa, Siobhan, and Gemma cited siblings 'back home' in the UK and Ireland who benefited from free childcare from their mothers, enabling full-time employment, socialising, and respite. Siobhan described how her mother in Ireland looked after her other grandchildren until they went to school, with her mother's house acting as a hub for the whole family. Her husband's family, and her dense network of friends stretching back to childhood, also live nearby. Her vision presents a stark contrast with the long days she now spends alone with her baby while her husband works long hours, six or seven days a week. She longs for someone to bring her a cup of tea or hold the baby for a moment. Even though she has never met up with anyone from the Facebook group, she finds it reassuring to know that she could. Observing online conversations reassures her that she is "not the only person feeling that way," that is, lonely and homesick.
Participants' sense of isolation was exacerbated by the sense that those around them (their husband, his family, and non-migrant friends) did not understand the emotional challenges they faced as migrant mothers. Finding others who shared their experiences empowered them to recognise the structural or contextual factors contributing to their troubles, which relieved their sense of personal failure. Winnie, who migrated from Malaysia to Australia © 2019 by Author/s 7 / 15 to marry an Anglo-Australian man, struggled for years to find friends. Her isolation was exacerbated by the lack of empathy in her partner relationship. Her husband, she reported, urged her to "try harder" to make friends, failing to understand the cultural and linguistic barriers, and racism, that Winnie encountered. When they moved from Queensland to Melbourne, Winnie joined the group for Malaysian mothers hoping to find "some quality friends that we can exchange heart." Winnie expressed a deep sense of relief on realising that other mothers in the group also found it difficult to make friends in Australia: Feel great, I feel [inhales] oh my goodness! After all these years, I finally feel that I find some friends. I'm not alone, and I'm not the one who has mental issues because I see everyone (…) has a common issue, in the sense of making friends in Australia.
Even women with migrant spouses who understood some of the issues were relieved to find a community of people who experienced similar challenges. Grainne explained that "it's like you're reassuring yourself that it's OK to feel this way; other people do." Katrin asked her German mothers' group about homesickness: "Is this normal? Will it ever go away?" The responses she received reassured her that it was a common experience and not a personal failure to adjust. Women who had become mothers after migration reported that motherhood had exacerbated their sense of homesickness. Gemma felt her Australian friends, while close, didn't understand her: "they're very close to us but I just feel sometimes that they don't fully understand our situation. Or how I feel sometimes." She compared them to the British and Irish mothers in her online group: "there's this supportive network of people who just understand, which is nice." Women did not need to participate in such discussions to gain emotional sustenance from them. Some declared themselves "too private" (Siobhan) or too "internal" (Lisa) to initiate online discussions but derived a sense of relief or solidarity from witnessing others' posts. Women understood the importance of responding, even just clicking 'like' on a comment, "just to show I understand, I hear you" (Siobhan). Merely observing these exchanges, in some cases simply knowing the group existed, produced a sense of solidarity, belonging and relief. "It is comforting to know that there are so many mums living in Sydney from where I'm from," Rebecca explained. Priya noted that, as a migrant, it was "comforting to know that you've got people around you." Women described their group as a "comfort cushion", a "safety net", and "a bit of security." The participants' words resonate with the literature around homesickness, which indicates that that "social support acts as a buffer against the negative effects of homesickness" (Scharp et al., 2016(Scharp et al., : 1192.
As well as providing emotional support for challenges common to many migrants, such as homesickness, the groups also helped migrant mothers manage challenges specific to their status as migrant mothers. Issues around migrant guilt and cultural transmission were experienced as gendered and linked explicitly to their position as migrant mothers. Women's migration-related guilt centred around a sense that they were depriving their children of extended family relationships, depriving their parents of close relationships with their grandchildren, were unable to carry out family responsibilities such as providing care for sick or elderly relatives, or had failed to spend enough time with relatives before they died. Kavita described herself as "completely wracked with guilt" during her first six years in Australia, guilt which intensified once she had a child. Women also expressed pre-emptive guilt about family emergencies or bereavements yet to occur. Grainne's mother-in-law in Ireland had had cancer the previous year and they had chosen not to return home but she stated that "if anything like that happened again, we probably would just go." Although Grainne's language implied that her husband shared her guilt and worry, a few minutes later she differentiated their responses: "I think if it was my mum, I'd have gone." Kavita made a similar distinction: If it had been me, and if it had been my parents, once I found out that there was a malignancy in the biopsy, I probably would have just dropped everything and gone straight home. But boys are quite different.
Although Wilding (2006) has suggested that online technologies have overcome some of the gender divide in transnational family matters, because men's increasing involvement via email relieves some of the 'kinwork' traditionally done by women (p. 135), research suggests that female migrants still feel more responsibility for transnational caregiving, particularly in-person or emotional care, and that migration-related guilt may be gendered (Baldassar, 2015;De Silva, 2017;Vermot, 2015). Migrant women's guilt, Baldassar suggests, is 'a ubiquitous and ever-present feeling of not having adequately met kinship obligations to care' (p. 87). In my research, mothers felt responsible for maintaining intergenerational relationships between grandparents and their grandchildren, facilitating frequent video calls, sharing photographs and making visits 'home'. These forms of kinship contact could not assuage their guilt at having disrupted those intergenerational relationships by migrating. Migrant mothers found little empathy from their spouses who, in most cases, did not experience the same intensity of guilt and often refused to participate in this digital 'second shift of affective and domestic labour' (Ouellette and Wilson,8 / 15 © 2019 by Author/s 2011: 559). Archana's husband, for example, refused to join the WhatsApp groups she had created to communicate with their families: "he's someone who doesn't want to be in all that." In response to this gendered imbalance in the affective labour and emotional load linked to their role as migrant mothers (and daughters), they sought empathy from women in the same position. For non-Anglophone or multilingual mothers, a desire to teach their children their heritage language was often a key motivation for joining the group, particularly those who faced incomprehension, indifference, or judgement from their Anglophone partner and his family. Women expressed a duty to pass on their language, similar to Manohar's concept of 'mothering for ethnicity' (2013a), whereby migrant mothers take on responsibility for ensuring their children grow up with a cultural identity that reflects their migrant background, as well as the key skills and knowledge that underpin such an identity. Mothers cited practical justifications for teaching their children their heritage language, such as communicating with family members, or increasing access to jobs and education. Similarly to the Australian-based Spanish-speaking mothers in Mejía's 2015 study, these mothers expressed a sense that their language 'is the most valuable thing that they can pass on to their children ' (p. 32). In an Australian context in which there is little institutional support for maintaining heritage languages, where they are seen more as a hindrance to migrant assimilation than an important cultural skill (Eisenchlas et al., 2013), responsibility for linguistic transmission falls wholly on migrant parents and communities.
Failure to pass on their language had an emotional impact that was difficult for those who were not migrant mothers to understand. For example, while Sabina noted that fluency in Swedish might facilitate her son's access to free tertiary education in Sweden, her main motivation was emotional, or identity-based. Sabina recalled her Anglophone husband's assessment of Swedish as "an insignificant language compared to Spanish," 13 to which Sabina declared, "you can't really say that about a language." Explaining the centrality of language to her identity, she said, "when I'm old and senile and have forgotten everything else, I'll speak Swedish." Teaching her son Swedish would give him a useful skill, which only she can provide, and it would establish a link between him and this core part of her identity. Many women expressed a strong desire to maintain a continuity between their own experiences and their children's experiences, and sharing a language was one element of this. The mothers' ambitions of raising bilingual children related to their aspiration to build a transnational identity (Utomo, 2014: 176) which would link them to their children through a shared language. Failure to transmit their language appeared to create a sense of rupture in their maternal narrative, eliciting feelings of sadness and loss.
The groups gave mothers access to advice on raising bilingual children, language-based playgroups set up by mothers in the group, and opportunities to maintain their own language proficiency. They provided solidarity and support for women encountering resistance or indifference towards this maternal project. Some described resistance from their children. "When I speak to [my son]," Katrin explained, "he's like 'nah, Mum, nah, I want to hear English, I don't want to hear German.'" Faced with their children's resistance, women pursued social relations with speakers of their first language to maintain the possibility of language transmission. As Eva stated hopefully, "we see Swedes every week, and at least it's there all the time." Some women's parents became "angry" (Annika) or "upset" (Katrin) at their grandchildren's refusal to speak their language of origin. This compounded women's feelings of guilt and failure. When Petra's son was excluded from the German playgroup, it felt like a personal failure, which her husband struggled to comprehend: Probably an Australian couldn't -like my husband struggles to [understand] why it was so important to me (…) He's like, come on, he will learn German anyway. (…) I felt, I tried to do this bilingual thing, and I kind of felt I'd failed or something. It was a sore spot.
In contrast to her husband's response, the German mothers' group understood the importance of language transmission to Petra and provided the emotional support she needed. Finding other migrant mothers committed to raising bilingual children in the face of Australia's 'persistent monolingual mindset' (Clyne, 2005: xi) validated their decision, and their identity as mothers with overseas heritage in Australia and supported their maternal linguistic project.
Establishing friendships is not a difficulty unique to migrant mothers, and in fact motherhood can, as discussed above, provide potential sources of new friendships. Nevertheless, the combination of these difficulties with the specific challenges faced by migrant mothers, such as time away from workplace sociality, the all-encompassing demands of young children, and their increased and unmet need for social support to help them establish and raise their families, exacerbated mothers' loneliness. In addition, the migrant mothers faced challenges around language transmission and the affective labour and emotional load that were specific to their position as migrant mothers. The online groups provided a source of support they were unable to find elsewhere. More broadly, the online groups make visible the existence of other people with whom they share important nodes of identity, past experiences, present challenges and future aspirations. The groups act as a material form of community, a means 9 / 15 by which they become visible to each other, can interact with each other and form relationships. In addition, the groups function metonymically, standing in for a wider imagined community who share their identity of origin, motherhood and migration. The existence of the group confirms that other mothers 'like them' exist, even if they are not all members of the specific online group.

MAPPING MULTIMODAL AND MULTILAYERED MIGRANT MATERNAL SOCIALITY
Based on the above analysis of the ways in which migrant mothers make use of online affordances, to obtain emotional sustenance from online interactions and to facilitate in-person social relations, this article proffers a tripartite framework of migrant maternal sociality that is multimodal and multilayered. In the interviews, women expressed the affective value of social interactions with other migrant mothers. That migrant maternal sociality can be differentiated into three levels of intimacy, all of which provide support, reassurance, companionship, and belonging. Firstly, women talked about casual contacts and interactions that functioned as entertainment. Secondly, they spoke about people who were neither family nor friends, but on whom they could rely for emotional and practical support. Finally, women spoke about the intimate friendships they had developed, or hoped to develop, from the group. In terms of modality, casual and intermediate interactions could take place online, offline or both, while intimate friendships were distinguished by increasing amounts of time spent together in person. Each mode of sociality contributes to an increased and multilayered sense of belonging: to a local community, an online community, and a transnational 'imagined community' of mothers drawing on common experiences and imaginaries of nationality, migration and motherhood. Distinguishing between these levels of sociality facilitates a nuanced analysis of the types of social relations and structures that have been disrupted by migration and motherhood, and of the different ways migrant mothers utilise the affordances of the Facebook groups to rebuild their social infrastructure.
Casual sociality could be dismissed as superficial yet the women spoke about it as a meaningful element in their social infrastructure. Migrant mothers felt isolated not only because they lacked intimate friends, but also because they lacked casual connections with whom they could gossip, exercise, go to the cinema, or whose quotidian dramas they could observe. New mothers missed being part of the sociality of the workplace, and wanted companionship during daylight hours when their friends were at work. While some women valued the ease of communication and common understanding that came from social interactions with people with a shared cultural identity or common language, for others that shared identity was simply a means by which they gained access to a potential social life. As Celine explained, It's just nice to know that you've got an invitation somewhere. So it's not that they are British, it's the fact that I've got an invitation and I happen to have got the invitation because I'm British.
Mothers felt their social lives had been severely curtailed by their child-raising responsibilities and lack of babysitters, so an invitation to a mothers-only gathering presented a rare opportunity to socialise without their children or partner. Even women who did not attend such events felt reassured that they could, if they wanted to, and enjoyed observing the plans forming in the online discussions.
Migrant mothers' groups present opportunities for light-hearted, playful chat, in online discussions and offline meet-ups. Although not necessarily intended to lead to deeper friendships, this chat was a significant form of culturally-inflected, everyday connection. Priya described the enjoyment of participating in online "chat or banter" with other Indian mothers, such as discussing plans for Valentine's Day, which she describes as "massive in India, even though it's not our festival." Her fellow administrator in the Indian mothers' group, Nisha, emphasised the importance of casual chat for encouraging women to participate in the group and thereby gain the confidence to ask for support on more serious topics. Susie recalled the "banter" from the offline British mothers' meet-up, discussing fashion, British clothes shops, and sourcing "real English chocolate" in Australia. Shared backgrounds make such banter possible. In turn, banter provides a familiar connection to home and culture that lends it more affective weight. As Francisco-Menchavez observes in her study of a Filipina migrants' community in New York, 'teasing, banter and laughter are just as important as the struggles' for producing what Manalansan calls 'diasporic intimacy' (Francisco-Menchavez, 2018: 115;Manalansan, 2005: 148). Even at this casual level, the groups are 'far more than a constellation of contacts and exchanges in social capital or a swapping of resources' (Francisco-Menchavez, 2018: 99).
At the other end of the spectrum of intimacy, the groups offered the possibility of close friendships, which many women had struggled to form since migrating to Australia or becoming mothers. When discussing whether people from the group had become 'friends', women usually mentioned meeting offline. "We're meeting up for coffee tomorrow" (Sherry); "We've done quite a few dinners and nights outs" (Eva). They also mentioned sharing important events such as children's birthdays, exchanging favours such as passing on children's clothes or cooking meals for each other when their husbands were away, or celebrating festivals together. Kavita imagined the future of a friendship from the group, as it might move up the scale of intimacy: We've done a few things, we've been to each other's kids' birthdays, our blokes have met, we'll probably do more things together. You know, you just kind of get to that stage where you know, as your network or your foundation, relationship develops, you get more overlap, and then it just becomes really natural, doesn't it, after a while?
These examples make visible the developmental nature of maternal friendships, as they move from online interactions to in-person exchanges, drawing their children and husbands along with them. Kavita's friendship with the other migrant mother helps to build their social infrastructure by drawing their families closer together. Participants were conscious of the work involved in moving friendships along a scale of intimacy. Becoming 'Facebook friends' was also cited as a step on the pathway towards 'real' friendship. As a performative act, becoming 'Facebook friends' can be a ritualised public display of a personal connection, involving rapid and intense self-disclosure (Chambers, 2013: 166;Lambert, 2013). By contrast, members of the same group can interact for months or years, while retaining close control over personal disclosures. As research suggests people interpret selfdisclosure online as representative of intimacy (Jiang et al., 2013), this move to Facebook friendship can be characterised as an escalation of intimacy.
Between the casual interactions and close friendships lies a thick level of intermediate relationships developed through the groups. Much of the significance of those intermediate relationships lies in the latent ties which underpin them. A latent tie, in this context, is a potential relationship of obligation, support, or empathy, derived from mutual membership of a group. It can be activated by a request for information or help but, before that request, has only a latent value. Haythornthwaite has established the concept of latent ties as connections which are 'technically possible but not yet activated socially ' (2005: 137), for example potential connections created by a departmental email list. The shared visibility of online interactions in the Facebook groups creates awareness of those latent ties, and what those ties might offer, providing a significant source of comfort for the groups' members. That comfort builds a sense of security through interdependence. Gemma described it as "instant (…) support from people who you don't know." Immediacy is particularly important to migrants who have not had the opportunity to build trusting relationships. The support could be activated at any time, as Petra explained: You don't have to be, like, deeply connected at all times, but when there's someone needs, there's a connection. You know, someone needs us now.
By developing these mechanisms to produce 'security through interdependence', migrant mothers create a counterbalance to norms of intensive and individualised motherhood, an emotionally absorbing and labourintensive mode of parenting for which mothers bear primary responsibility (Hays, 1996;Henderson et al., 2016). This maternal responsibility can feel even more intense in the context of migrant isolation. Members of one Indian mothers' Facebook group helped a newly separated mother find culturally appropriate accommodation. Another group delivered meals to a struggling mother who had contacted the administrator for help. Brazilian mothers helped find employment for members' husbands who had lost the jobs on which their family's visa depended. Locked out of her apartment with her two young children, Priya posted on the group and received instant offers of help. Drawing on their membership of the group, in turn based on mutual recognition of each other as mothers and migrants from similar geographical or cultural locations, women felt able to seek help from each other. Women observed as their fellow mothers offered and received tangible support in times of need, and felt comforted, even if they did not participate any further: It's just a virtual connection for me at the moment because I've not been to any meetings but I do want to. It feels like you've got a little fall-back, in a way, you have got that safety net, you know, if you are having a particularly hard time, you've got a safety net you can go to. (Gemma) A shared identity as migrant mothers from a particular background intersects with the "mediated intimacy" (Chambers, 2013) of online sociality to create a glocalised maternal safety net consisting of latent, intermediate ties.
In Sabina's words, "It's like a fire extinguisher -you might never use it, but you have it there. So you're happy it's there, right?"

CONCLUSION
Migrant mothers, at the intersection of maternal isolation and migrant loneliness, use glocalised online maternal communities to rebuild their multiply disrupted social infrastructure. This reconstructed infrastructure encompasses casual, intermediate, and emotionally intimate forms of sociality. Each of these contributes to an increased and multilayered sense of belonging: to a local community, an online community, and a transnational 'imagined community' of mothers drawing on common experiences and imaginaries of nationality, migration and motherhood. These senses of belonging draw on attachments that are place-based, identity-based, experiential, and relational. Through these varying levels of social intimacy, migrant mothers gain companionship, reassurance, comfort, and belonging. In contrast to claims of a crisis of intimacy resulting from a reliance on technology and social media for social interactions (Turkle, 2011), analysis of this data suggests that migrant mothers use the affordances of social media to create and access forms of intimacy, support and belonging that are necessary to their wellbeing, and which their existing networks of family and friends cannot provide.
In these communities, the global and the local are mutually constitutive, shaping women's experiences of the maternal. Through their membership of these glocalised online communities, women position themselves within imagined communities of mothers whose experiences, practices and social relations carry significant affective weight. In contrast to the close-knit 'communities of care' in Francisco-Menchavez's research with Filipina migrant domestic workers in New York, in which 'fictive kin' relationships develop out of shared experiences of trauma, the online migrant maternal groups in this study should be understood as looser collectivities providing multilayered opportunities for sociality and emotional support. These opportunities are facilitated by the online groups but may take place either online or offline. Friendships that move more firmly into an in-person mode are characterised as more intimate than those that remain wholly online. However, the groups' online modality makes possible 'security through interdependence' through the creation of latent ties. This sense of security and readily available empathy and support provides emotional sustenance for women experiencing the challenges of migrant motherhood. Many women attested to this emotional benefit, even when their involvement was limited to observing the interactions of others in the group. The online modality of the groups enables women to participate at a level that meets their needs, whether that involves observing other mothers' interactions, seeking help or responding to others' requests.
This multilayered and multimodal framework of sociality has been developed out of the field research with migrant mothers, and provides insights into the mechanisms for 'horizontal caring' migrant mothers create, using the affordances of Facebook. The disaggregation of the different layers of sociality, and the identification of the emotional significance of the latent ties created by the ability to witness exchanges of support, might also be usefully applied to other contexts in which digitally literate but isolated people need to (re)build a locally-situated social infrastructure, for example university students leaving home or military families.
A focus on migrant mothers' relations with each other, and their potential to form an affective resource for each other, offers a perspective which honours their selfhood beyond their maternal identity and emphasises the importance of lateral relationships between mothers. 21 st -century feminism would do well to attend to the significance of these sustaining relationships between mothers, particularly in the challenging context of migration.