Techno-emotional mediations of transnational intimacy: social media and care relations in long-distance Romanian families

The transnational circulation of intimate care is increasingly mediated by digital communications. Research conceptualizing long-distance emotional intimacy in terms of ‘care chains’ has been influential in understanding international care economies. Yet, this framework has limitations for theorizing the role of media in communications of care. With a focus on the ‘left-behind’ family members of Romanian economic migrants, this paper investigates how the use of social media apps and mobile devices within the context of a major intra-EU labor migration phenomenon helps people stay in touch with their transnationally mobile loved ones. It draws on interview material elicited among the close family members of Romanian labor migrants living in Bucharest and surrounding areas. The analysis focuses on the sensory role of social media platforms and the materiality of smartphones in shaping relations of long-distance emotional care. Showing how video calling and photo sharing practices produce emotional experiences that are specific to contemporary combinations of platform-device technicity and social sensitization, the paper argues for conceptualizing transnational care as a mediated emotional experience. By theorizing the role of media in how care is not merely transferred but felt through mediation, the paper demonstrates how media practices produce a techno-emotional mediation of transnational care.


Introduction
For contemporary transnational families, living at a distance from loved ones means emotional intimacy is increasingly shared through social media platforms on a routine basis. This condition raises questions about the dynamics of emotional care between transnational migrants and their loved ones as platforms come to play a growing role in people's everyday communications (Acerdera et al., 2018;Baldassar and Merla, 2013;Baldassar et al., 2016;Nedelcu, 2017). Such questions have emerged within a relatively new research area on digital media and transnational migrants' emotional lives (Baldassar et al., 2016;King-O'riain, 2015;Madianou, 2016;Madianou and Miller, 2012;McKay, 2012;Nedelcu and Wyss, 2016;Panagakos and Horst, 2006;Wilding, 2006;Witteborn, 2014). As Holmes and Wilding signal, intensified spatial mobilities in line with the demands of the global labor market -be they long-term transnational migrations or short-stay, commuter, or seasonal mobilities -have changed social relations such that we require 'new frameworks for thinking about how people live their social and emotional lives', including through online spaces such as Facebook, Instagram, and WeChat (Holmes and Wilding, 2019).
Concepts recently introduced for understanding transnational family relationships in the digital age include 'digital kinning' (Wilding et. al., 2020;Baldassar and Wilding, 2020), 'doing family practices' (Nedelcu, 2017;Nedelcu and Wyss, 2016), 'ambient copresence' (Madianou, 2016), 'digital kinship' (Hjorth et al., 2015), 'mediated co-presence' (Cabalquinto, 2019), and 'careful co-presence' (Alinejad, 2019). These have largely converged around the idea that migrant family relationships are produced through repeated, everyday activities of enacting social ties. Overall, this convergence has consolidated a wider shift away from a notion of kin as a normative category of relationships, and toward an idea of family relationships as (re)produced through people's active emotional 'doings'. Aligning with wider theorizations of 'kinwork' done to maintain families (Di Leonardo, 1987), these doings are understood as activities increasingly undertaken through social media apps. The idea of family relations being maintained through effortful activities (of women) overlaps with important discussions of transnational care as 'work' or 'labor'. Indeed, the notion of intimate care as emotional work has been compellingly conceptualized in literature on migrant care in terms of 'chains' (Hochschild, 2000;Parreñas, 2001) or 'circuits' (Baldassar and Merla, 2013). This frame has been imperative for rendering gendered (migrant) labor a visible part of the global economy. The literature on 'doing family' digitally has also tended to highlight the important ways in which social roles in families are maintained with relation to (gender-) normative expectations of who must do what kind of 'emotion work' under conditions of social media's proliferation (Sinanan, 2019). Yet, the exploitative associations of 'labor' may obscure ways in which (women's) care activities may become coercive, undertaken to gain control, or to create obligations in others in the household, as some have observed (Di Leonardo, 1987;Drazin, 2011).
Moreover, an understanding of the doings of transnational care as 'labor' renders some of the important experiential dimensions of care peripheral to the discussion, and conflates various kinds of activity. I argue that foregrounding the experiential qualities of emotional care and focusing on specific modes of doing expands possibilities for further theorizing how care is technologically mediated. Through an investigation of how family members who stayed 'behind' in Romania use digital media to communicate with their loved ones abroad, I address this paper's central question of how transnational care is digitally mediated, suggesting how media scholarship might further enhance understandings of care in contemporary contexts of globalization. I conceptualize the long-distance emotional experiences I present in this paper as the technoemotional mediation of care, drawing on interview material gathered in Romania among the family members of those who emigrated from Romania under various circumstances. Romania has not only seen recent concomitant booms in both intra-EU migration and smartphone use, it also has a post-soviet history of massive domestic care labor migration to Southern European countries.
In what follows, the first section outlines conceptual discussions of emotion, media, and transnational mobility in the fields of migration and diaspora studies, media studies, and digital anthropology, bringing debates on emotion from media and internet studies into conversation with the work from migration studies and (digitally mediated) transnational care. The second section goes on to give background to the empirical investigation of the Romanian family members who stayed behind, 1 as well as explain the research methodology. In the third section, I argue, on the basis of interview material, that the visual affordances of photo-sharing and video-calling cultivate techno-emotional ways of looking with care. Such looking involves feeling care through its platform-mediated expression; a 'doing' that integrates media use with bodily practice and sensation to produce emotional experience. The fourth section's analysis builds on these claims by showing how ways of looking with care are facilitated not only by technology use, but also by care mediators whose practices normatively shape how care is mediated. I conclude by discussing the implications of this argument on emotional mediation for conceptualizing experiences of transnational care.

Emotional mediation and transnational mobility
Debates about emotion/affect and digital media have recently burgeoned within the fields of media and internet studies (e.g. Bareither, 2017;Chambers, 2013;Döveling et al., 2018;Garde-Hansen and Gorton, 2013;Giaxoglou and Döveling, 2018;Grossberg, 2010;Hillis et al., 2015;Karatzogianni and Kuntsman, 2012;McVeigh-Schultz and Baym, 2015). With regard to the role of social media, Papacharissi's influential notion of 'affective publics' engages with the question of how affect takes particular shapes within platform spaces (Papacharissi, 2015). Focusing on Twitter, Papacharissi argues for the formative role of affect in producing publics through networked connections. In doing so, she makes the case for both the integral role of affect (for Papacharissi, distinct from, but necessary for, feeling emotion) in social and political dynamics, and the importance of investigating these dynamics through the web-based platforms through which mass publics are increasingly formed. Following on from the idea of platforms giving their own tonality, texture, and rhythm to the public events they mediate (Papacharissi, 2016), it seems equally important to ask how platforms may change the subtle resonances of intimacy, closeness, and privateness. How are these felt, registered, remembered, and given meaning in the context of people's digitally mediated interpersonal relations?
The 'mediatization of emotion' framework  has also been advanced to understand emotion on social media platforms, suggesting that a 'digital affect culture' (seeing affect as constitutive for emotion) helps explain how emotion is communicated or 'done' within digital cultural environments at three scales of intimacy (what are referred to as the intersecting levels of the micro, meso, and macro). While these ideas help scholars argue for the importance of the emotional dimension of digital communications, the mediatization of emotion also suggests that 'the digital terrain' has a different 'logic' to its 'emotional flows' than 'the flows outside the digital realm' (Döveling et al., 2018: 7). However, given that digital anthropologists have problematized the idea of built-in platform logics in favor of a notion of 'affordances' that emerge through digital media practices (e.g. Costa, 2018), the idea of a digital logic of emotional spread warrants further scrutiny.
Media anthropologist, Bareither, draws on a practice theory of emotion and media to argue that digital platforms have particular 'emotional affordances' (2019) that emerge through people's usage practices to produce emotional experiences. Drawing on the idea of media practices producing experience, others have shown how social media practices produce mediated transnational experiences of long-distance homeland connections, and emotional intimacy with family (Costa and Alinejad, 2020). These approaches to media practices in contexts of migrancy support the assertion that media is 'an irreducible constituent of the social experience of mobility itself' (Keightley and Reading, 2014), making it all the more relevant to investigate these mediations in relationships marked by migration.
The spatiality of transnational mobility and distance is relevant for the study of close emotional relationships, as Holmes and Wilding (2019) have argued. Emotional closeness has within it a metaphor for spatial proximity, as the scholarly discussion around the concept of intimate 'co-presence' in the digital age has aimed to unpack (Ito, 2005;Madianou, 2016;Nedelcu, 2017). The spatial dimensions of the body's situated-ness in the world is of compound importance when considering digitally mediated emotion in migration contexts, especially given the significance of geographic distance/proximity, embodied use of technological artefacts, and potentiality of being emotionally affected (Alinejad and Olivieri, 2020). This claim draws on the argument that while emotion is given meaning, culturally, the body is always its site (Gammerl et al., 2017). And while emotion circulates spatially, it is also rooted in place (Navaro-Yashin, 2009). This investigation's ethnographically-informed approach to social media platform usage therefore aims to access migrants' everyday, situated ways of living emotional lives across distances and borders, by looking at people's ordinary practices of distant communication via social media platforms. In doing so, it acknowledges, as Keightley and Reading (2014) have pointed out, that while it is important to theorize the relationship between media and mobility from the perspective of people's lived realities, the role of emotion has been underacknowledged in such theorization.
The case of Romanian transnational families brings particular forms of intra-EU migrant mobility to the center of the paper's discussion of emotion and digital media. The analysis starts from details of users' socio-spatial ways of living emotional lives in transnational families, and in the cultural contexts of their local settings -these extend to sites of web use outside the wealthy Western world and capital cities, with the field site being located within but also beyond the city of Bucharest. In the following, I elaborate on the details and significance of the field case.

A changing context of migration and mobile web device use
The empirical focus of this paper is Romanian economic migration to European core (or less peripheral) member states. 2 Romanian migration has attracted scholarly attention around questions of care and intimate transnational relationships, with an emphasis on family and the parent-child dyad due to historically gendered migration patterns (Monini, 2018). However, in contemporary Romania, migration is rapidly diversifying due to EU border policy developments in recent decades. 3 Hence, the respondents whose stories around which this paper is developed include family members of domestic workers who migrated in historic migration flows, as well as those whose intra-EU mobility has intensified with Romania's more recent EU integration (these include circular migrants and seasonal labor migrants in sectors such as construction and agriculture).
While Romania has among the lowest rates of daily internet use of all EU member states, 4 it has also undergone a striking recent surge in internet use on mobile devices, with this rate more than tripling between 2012 and 2017. 5 Contemporary Romania therefore complicates a static/simple notion of a geographic 'digital divide' by exhibiting both a relative lack (in comparison to EU) and abundance (in comparison to its own recent past) of networked communications. Scholarship on Romanian migrants' digital communications has argued that everyday digital communications produce ways of maintaining relationships that form cosmopolitan subjectivities (Nedelcu and Wyss, 2016) and diasporic political identities (Trandafoiu, 2013). While this scholarly attention has tended to go to (middle class) migrants (see Ducu, 2018 for multi-sited research), the communication practices of those 'left behind' in the home country by their migrating loved ones are an understudied but equally important part of transnational families' communication and care dynamics (Cabanes and Acedera, 2012). Hence, this paper focuses on those who remained in Romania to show what an examination of mediated experience might reveal about how contemporary transnational care is 'done'. In what follows, I further describe how the field research with these respondents was conducted.

Methodology
In the summer of 2018, I conducted short-term ethnographic fieldwork (Pink and Morgan, 2013) on social media and smartphone use among a set of 20 Romanians 6 (in an even gender split) who had experience with their family members migrating for work to wealthier European countries. The investigation focused specifically on respondents' use of social media to communicate with close family members. While the short-term nature of the research also meant that the investigation was primarily interview-based (rather than involving immersive participant observation), the interviews were conducted in the homes of respondents or in places they frequented often, hence the details of settings were also observed first-hand and recorded, as were instances of device and app usage. Each of my respondents had close emotional relationships with people who were economic migrants in European countries outside Romania (Germany, Spain, Italy, Greece, and UK). These included immediate (biologically related) family members such as parents and children, but also some siblings and a few romantic partners. I took my respondents' mediated practices of care and intimacy as the unit of analysis, regardless of whom they had close relations with. I selected the sample through a combination of entry points through NGO gatekeepers and other contacts gained locally and internationally in the course of preparing for fieldwork and living in Bucharest.
These entry points then led to other respondents from within the same close social circles or families through the snowball method of gaining introductions to further respondents via other respondents. Hence, small clusters of respondents within the sample knew one another. The theoretical sampling goal was to include as much diversity in the sample of respondents as possible, specifically along class and ethnic lines, in order to be able to understand what might be particular to and common across people's communication practices under a range of circumstances. The forms of labor migration in question ranged between seasonal, unskilled labor to longer-term skilled work. And there was an approximate even split between those who were middle class and educated versus those who had less education (although in several cases of upward economic mobility among the younger people, education did not align with the socio-economic background or skill level of the parents and extended family), and approximately half the respondents were of the Roma ethnic minority. Ages ranged from 20 to 65 years in order to include the adult children of migrant badantes who migrated decades ago, the elderly carers of young children, and those with adult siblings abroad. Each formal interview took approximately 1 hour.
In cases were NGO organizations were the entry points, NGO workers acted as mediators helping to establish rapport, translating from Romanian to English in instances where this was necessary, and giving additional background information about the family constellation. While the help of key respondents and institutional gatekeepers was crucial in facilitating access and communication, it also defined the terms of the interaction in and around the interview setting rather strictly. Hence, I am cognizant of how the varying degrees of cultural/linguistic commonality between my respondents and myself led to discrepant levels of depth with which I could relate to their narrated experiences. 7 However, the fieldwork involved spending longer amounts of time with my gatekeepers/ cultural and linguistic mediators -meeting them on numerous occasions and having them mediate multiple interviews each -which facilitated the elicitation of contextual cues that helped me to understand more about their specific biases and expertise, something I believe informed more sound interpretations of respondents' accounts on my part.
Furthermore, given the specific research focus (on social media practices of communication with loved ones abroad), the depth of investigation was more than sufficient to yield material from which I was able to identify relevant patterns of practices on the basis of which to analyze social processes. These were analyzed through an iterative combination of open coding, which allowed important themes to emerge from the interview transcripts, and more structured coding with relation to the central research themes of emotion, care, and long-distance intimacy. I acknowledge that the short research period and translation practices produce limitations to the depth of the material gathered, while I also rigorously analyze this material for what I think it does show, namely how people make sense of their conditions of distance with relation to their loved ones and how they narrate the typical ways in which their uses of digital media platforms and devices feature in their transnational family relationships. I believe the advantages of accessing perspectives that would otherwise have been omitted without translation and mediation outweighed the abovementioned limitations, a claim to which I hope my following arguments attest.

The technicity of looking with love
Photos and videos were an important part of how family members in Romania communicated with their loved ones abroad. A recurring theme I noticed in people's descriptions of their photo sharing and video calling practices on social media apps was the insight they considered these visual media forms to offer them into the emotional states of the person they were talking to. For example, Marianna was 57 and had two sons in their 30s living abroad -in England and Germany, respectively. She said that she liked to use video calls through Facebook Messenger to see her children, and she spoke to them more now than when they were in Romania: Because, you know, they're among strangers now, so I'm a bit more worried. I'm their mother, so I need to know if they've eaten, if they have clean clothes, if they're fine, you know, how they're feeling. They also need to know about me because I'm ill. So, they need to make sure I'm fine as well.
[. . .] It calms me down to just see their faces and hear their voices myself. They show me the room where they live, they assure me they have a fridge with food in it, and it's reassuring because I can see it all, so I know they're not lying to me.
Marianna says she tries not to cry when talking to her sons via video calls. She finds the calls difficult sometimes yet they also give her peace of mind seeing with her own eyes that 'they are fine'. In an analysis of the relationship of truth to care in transnational families, Baldassar (2007) emphasizes the status of seeing with one's own eyes as a measure of truth. While her work refers to 'seeing' as visits involving physical presence, a similar regime of truth can apparently be extended to the seeing afforded in video calls, at least when it comes to people's experience of accessing their distant loved ones' true emotional state.
The multiple accounts like Marianna's I encountered suggested it was not the mere use of video image in these calls that made reading of emotions possible, but a particular way in which people looked at the visual content. Stefan's example illustrated this most evocatively. A university student in his early 20s, Stefan had an older brother living abroad. Their bond had always been close, and his daily life at home had been affected by his brother's absence. But he was soothed to be able to see the check-ins and photos his brother posted on Facebook periodically, specifically because his particular way of looking at them gave him the feeling he had access to his brother's emotional state.
I can see if he smiles a true smile, or he's trying to smile for photos. So I see how he feels over there, emotionally.
[. . .] Another person will see a smile, but I can see if it's his true smile or that he tried to smile and it's forced for the photo.
[. . .] Sometimes I even make a zoom of his smile to see it better.
Stefan's intimate relationship with his brother allowed him a way of looking at his brother's photos that made it possible for him to feel he had access his brother's emotional life. His emotionally invested practice of looking, together with the technical affordance of zooming into digital photos on his smartphone by pinching the touchscreen on his device, reveals social media practices beyond sending, sharing, receiving, liking, and scrolling. Rather, an emotionally sensitized mode of looking at periodically shared social media content on personal smartphone screens can be seen as a habitual sensorial engagement that cultivates an intimate attunement with the distant family member through the particular media form. Similar processes of sensitization have been happening since people started listening in particularly attuned ways to their loved ones' voices, as Baldassar (2007) compellingly describes in her discussion of how migrants manage truth and distance, together. But Marianna and Stefan's practices suggest modes of looking that do not isolate looking in the way that phone calls isolate listening. They involve the interactive audiovisual communication of video calls, and the tactile looking that the software-hardware assemblage of the smartphone touchscreen offers. This supports Pink's argument about media and the senses, which emphasizes the integration of all the senses within the bodily sensorium, and states that new media forms do not make people's experiences more sensorial, but that 'new ways of touching, looking, and moving with digital media . . . have alerted us to the sensoriality of our embodied and affective engagements with media in new ways' (Pink, 2015: 6).
This mode of engagement with social media gives the respondents left behind in Romania the sense that they have an up-to-date insight into the emotional condition of their loved ones abroad. The importance of this sense takes on specific significance here, as the purpose of migration is to secure an improved standard of living for the migrant, those left behind, or both -something many of my respondents spoke about in terms of 'having a good/better life'. Checking on loved ones' emotional state, therefore, is not only related to the intimate dynamics of family relationships over long distances but also about communicating (or concealing communication of) how the conditions of everyday life on both sides of the communication are affecting people. Looking with love at photo and video content therefore not only produces the emotional experience of being in a close relationship, but also shapes people's feelings about the condition of the transnational migration that marks their everyday life. The choice of the loved one's migration is therefore constantly assessed and reassessed indirectly through the routine checking-in on their emotional state at a distance, in a seemingly ongoing reflection process of the overall benefit or necessity of the act of migration. This raises the stakes of authenticating others' feelings from a distance. As cases like Marianna's and Stefan's suggest, caring ways of looking at mediated visual images produce experiences of emotional reassurance for those doing the looking.
In their constant communicative assessment, sometimes people come to feel the costs of migration outweigh its benefits. Andrea said that she developed an emotional bond with her daughter for the first time through video calls. Just a few months after their daughter's birth, Andrea and her boyfriend both left Romania to work in the UK. He worked in construction and she worked in a bar in London. She recalled that early on, she would sometimes forget to call her daughter in the evenings because she had too much on her plate at work, finding out that she was too late because her parents had already put the child to bed. Wanting to work and live in London with her boyfriend and finding the choice between staying and return difficult, Andrea said she devised a way to alleviate the guilt of not feeling the urge to call her daughter by gradually building a relationship with her via video calls. She explained: I was just watching her for ages. I mean, even if she was asleep, I was just looking at her. That was enough for me then. Even when I was at work sometimes, I would just leave the video call on, because I was behind the bar. So it was easy for me just to have a look. And I just put the phone under the sink somewhere, so I was like, I'm just doing my things and at the same time watching her. That was a cute thing. It helped me, to be honest.
After a while, Andrea came back to Romania to raise her daughter. Now that her daughter was 3 years old, Andrea feels she could never imagine leaving her to work in London anymore due to the bond that had developed over these past few years. While the affordances of social media apps for constant intimate contact in transnational families are well documented, Andrea's case shows that social media communications do not only become background elements to which users grow accustomed. Rather, particular emotional attunements seem to be cultivated through routines of looking. Andrea's desire to return home to look after the child she had left in the care of her parents is a fulfillment of her normative social role as a mother, as was perhaps her choice to buy an iPad to serve as a baby-monitor. However, her narrative suggests that it was not her obligations but the experience of the long-distance sessions of watching her baby by video that eventually moved her. Andrea decided to move back to Tecuci with her parents and daughter, while her husband continued to travel back and forth between Romania and Italy, where he had since set up a construction company.
While the photo sharing and video calling practices I have discussed here have different technical affordances with regard to historicity/instantaneity and sensory combinations, what I find relevant are the commonalities in the sensibilities with which both kinds of visual images are looked at. This suggests that platform-device assemblages are taken up in specific ways within relationships of intimate care. The emotionally caring mode of looking discussed in this section is a 'doing' through sensory engagement that can be distinguished analytically from the media practice of platform usage defined by scrolling, sharing, or liking. It is a sensory mode of doing that is defined not merely by the platform's design and functionality but also by the emotional relations of care it is being used to mediate. Furthermore, the emergence of a mode of looking with care as a media practice that produces an experience of reassurance or longing for the looker suggests that care addresses the emotional needs of the carers, themselves. Hence, care does not only circulate here because there are unequal power relations that compel carers to occupy servile positions of caring for others, but because care is a way for those who 'do' it to seek out and feel their own feelings about their condition within the relationship. In what follows, I discuss further how caring modes of sensing are cultivated through media practices, acknowledging the significance of what has been called 'care by proxy' (Baldassar, 2014).

Care mediators and social sensitization
All instances of children being left behind by one or both migrant parents that I encountered involved an extended network of family members who not only became the ones to take care of the immediate needs of the child(ren), but also invariably came to take on the role of brokering the communications between the parent(s) and the under-aged child(ren) to some degree. When guardians described how they helped children communicate with their parents, and when the adult children of migrants recounted memories of their childhoods, the deep significance of these mediators in facilitating the use of various media forms for transnational communication became readily apparent. These caregivers were not only implicated in transnational family units. As I argue in this section, they also played a crucial role in producing emotional sensitization through social media apps of the kind I described in the previous section. In the cases I observed, when the father was left behind, the paternal family stepped in. If both parents migrated, the maternal grandmother became the primary caregiver. Across the instances of children staying behind with other family members while one or more parents undertook long-term migration abroad, a pattern emerged as to the way communications between the children and the parents were remembered and changed over time. Namely, the image of the (usually, but not always, maternal) grandmother who gathers the children around in the living room and uses the landline to place a long distance phone call to the parent(s) was a repeatedly recounted early memory.
Typically, the children would each have their turn on the analog dial phone after the grandmother had opened the exchange. As the corded receiver changed hands, the children would each speak to their parent with the audience of the others in the room. All respondents with their own variations of this recollection spoke about the mediator as doing the work of fostering the relationship between parent(s) and children, whether or not these efforts resulted in a sustained relationship. These mediation activities involved practically arranging phone calls between parents and children, but they also included engaging the children in the communication or covering for parents if they became unavailable. These mediation practices remained important for reconciliation and reframing feelings of hurt or neglect, even as children grew into young adulthood and replaced the living room landline with their own smartphones. But at times this mediation also inadvertently contributed to feelings of disappointment and exclusion from care when communication was truncated or partial. For example, Corina had received permission and money to buy a smartphone at an early age. At the time, she noticed this was something none of her other classmates, whose parents lived in Romania with them, had. In most cases I encountered, a smartphone for children who stayed behind was one of the important first purchases parents made. An early adopter who had been raised by her maternal aunt, Corina was now a university student living in Bucharest, and her mother, with whom she had a strong bond, had remained a care worker for various employers in Italy for over two decades. Corina's account of a recent conflict between herself and her mother was striking, both for how her aunt had tried to help mediate the communication between mother and daughter, and for how this had (re)sensitized Corina to voice calls versus video calls on WhatsApp.
There had been a period when Corina's mother chose not to video call her daughter on WhatsApp, but instead used voice-calls. Corina found this strange and it made her uneasy, so she asked her aunt if anything was wrong. Her aunt reassured her that nothing was meant by this change, but Corina later found out that her mother had been experiencing economic hardships during this period due to employment troubles. This had forced her to lose her apartment and she had ended up staying on the couch of a friend for a period. During this time, Corina's mother had avoided video-calls to prevent making her difficult situation visible, and had asked Corina's aunt to keep the bad news from her. Upon learning the whole story, Corina was deeply hurt about being kept in the dark during a period when she could have offered emotional support. While she said she had forgiven her mother, she was still not speaking to her aunt over this incident for the past weeks. She said she still relived the uneasiness of the voice-calls each time her mother called -in the split second it took to make sure that the call was a video-call, she was reminded of her past negative experience.
Despite being marked by their own hardships, accounts such as Corina's were ones in which parents had maintained close bonds with their children through sustained communication and the crucial efforts of mediators who were from within the family. While the omission of bad news in these cases was meant to reassure those left behind, it had instead cultivated in them a keen, distrusting sensitivity to omissions and silences. In Corina's case this sensitivity was bound to the voice call without image, and others had similarly described feeling distrust toward the choice of their loved one to switch to voice calling from video calls. Yet, these cases were ones in which mediators tried to assuage this sense of disappointment or emotional distrust. And in this sense, such instances were dissimilar to cases where migrant parents could be said to have abandoned their children. The instances I encountered of parents engaging in only rare, brief, and unreliable contact with their children from outside the country emerged in families where life under conditions of considerable poverty was compounded by other factors such as parents' and/or grandparents' lacks in literacy and/ or primary education, disabilities, or involvement with criminality. These situations were few in my sample, but they made sharply apparent that the mere normative social role of motherhood, together with the possibility of long-distance communication with the child, was not sufficient for relations of care to emerge in the family. In such cases, the circumstances were sensitive and it was not entirely clear why the parent had left and cut most contact. In this absence, mediators did more to take over and spread the responsibilities of the parent within the present family members and social services than to try to sensitize the child/children to the long-distance care of the parent.
For example, Nadia was 20 years old and had been cared for and supported by her grandmother, her aunt, and social workers since her mother had left to Greece when Nadia was a toddler. After making phone calls to her family for the first few months, Nadia's mother sent a few packages with books and other gifts but later they also stopped, as did any remittances. Nadia now spoke to her mother only on very rare occasions and told me plainly that she knew her mother was living with her boyfriend in Greece, not working, and had no plans of coming back to Romania. Nadia's social worker and Nadia herself explained that no amount of mediation on the part of other family members was able to convince her mother that she should take up her full role as a parent to Nadia. Instead, Nadia's carers encouraged her to focus on her studies and her extracurricular hobbies and her relationship with her local relatives. Her everyday use of social media platforms was directed mostly toward communicating via Snapchat with her cousins, the children of her mother's brother, living in the UK.
Migrant parents leaving their children in Romania relied on the care of others. In by far most cases, these local mediators were family members who also stayed behind in Romania. Depending on the circumstances, these mediators of care shaped the experience of those who stayed behind, whether this was by sensitizing young people to the long-distance care of their parents, helping parents conceal bad news to avoid spreading worry, or managing emotional responses to the (perceived) lacking care of parents. Developing the argument made in the previous section about how mediated experiences of care are shaped by the affordances of the user-technology sensory interface, here I highlight how that interface never takes shape in a social vacuum but is directly influenced by other family members. This influence evidently features the force of social roles according to inter-generational, life cycle stage, and gender norms, even if expectations of what these entail appear to be continuously negotiated. In fact, these social norms seem to gain or lack force according to emotional sensitization processes in which normative moral responsibilities and entitlements come to be felt with more or less intensity. This is evidence for the role of emotion in the ways people engage with platformmediated communications of care. And the role of family members who mediate long-distance care communications is less one of a care 'proxy' and more a part of the process of care mediation that attunes the senses more or less toward certain messages or others.

Conclusion
This paper has suggested that approaching transnational care in Romanian families as an emotional experience that is platform-mediated helps us better understand the formative role of emerging technologies in distant care relations. The device-software assemblages required for social media platform usage appear to produce the experiences of long-distance emotional care necessary for relationship maintenance rather than simply being a conduit for the circulation of care. The cases discussed demonstrate how the technological assemblages of hardware and software become implicated within a techno-emotional process of mediating transnational care, a process by which platform-device technicities, embodied media practices, and normative social forces together give rise to intimate modes of sensory (dis)engagement. This insight builds on perspectives that point to the convergence between digital technologies and parts of the physical body within assemblages (Wise, 2012), offering a materialist perspective on understanding communication. Such perspectives help attune our understanding of mediated care to the often ignored habits and routines through which feelings come to be managed or governed, sensitized or dulled, regardless of whether these involve media technologies, bodily practices, social surroundings, or a combination of them all.
From the interview material gathered about people's platform uses under a variety of conditions in Romania, it emerged that indeed some were charged with gendered social roles that called upon them to exact more activities of initiating communications, foster care relationships between others, and take on other care responsibilities. However, the forms of sensorial doing people engaged in were also motivated and shaped to some extent by their own emotional needs and experiences. In this paper, I have suggested that understanding transnational care as a mediated experience helps to move beyond some of the limitations that the labor-oriented analytical frame that has been extensively applied to both digital activity and emotional practices. This argument implies a need to not only challenge our understandings of contemporary transnational emotional care to better specify what the 'doings' of contemporary care-at-a-distance entail beyond the notion of 'work', but also to better explain where care comes from, and how and why it comes to be felt (or not).

Funding
The author(s) disclose receipt of the following financial support for the research, authorship, and/ or publication of this article: the ERC (European Research Council) consolidator grant "Digital Crossings in Europe: Gender, Diaspora and Belonging" (CONNECTINGEUROPE) (Grant Number 647737).