Power triangle: Power dynamics and children’s agency in Chinese liushou (left-behind) families

Abstract

This study explores power dynamics in Chinese liushou (left-behind) families from the perspective of children. Drawing on 9 months of ethnographic fieldwork in rural Sichuan, it demonstrates that intergenerational authority is reconfigured into a “power triangle” linking migrant parents, co-resident grandparents, and children. Children’s agency emerges not only through forms of resistance such as concealment, humouring, and selective compliance, but also through retreat, shaped by guilt, sacrifice, and compassion. The analysis conceptualises power as ambivalent, intimate, and negotiated, offering new insights into how migration reshapes family life and generational orders.

Introduction

In the era of China’s great migration, more than 66.93 million children live apart from their parents, who migrate to earn a living for the family; among them, an estimated 41.77 million reside in rural areas under the care of a co-resident parent or grandparents (UNICEF, National Bureau of Statistics of China, and UNFPA, 2023). Commonly referred to as liushou, these children remain in their hometowns while parents work elsewhere. Unlike the English term left-behind, liushou conveys the idea of “staying and holding the fort” (Xiang, 2007: 6), reflecting a cultural expectation that family members continue everyday life and maintain the household while awaiting migrants’ return. In this article, we use liushou to foreground its Indigenous meanings and the cultural expectations that structure Chinese family life (Guo and Spyrou, 2024).

Parental migration has significantly changed family structures and care practices (Ye and Pan, 2011). To adapt to these changes, different family members, including liushou children, their migrant parents, and substitute caregivers, fulfil relational obligations to one another as a form of mutual commitment (Murphy, 2020), echoing findings from worldwide migration studies (Dreby, 2010; Lam and Yeoh, 2019; Punch, 2007). Beyond these structural adaptations, however, migration has also interacted with broader cultural shifts in how family relations and obligations are understood. In contemporary China, marked by the transformation of filial piety and the rise of increasingly child-oriented families (Fong, 2004; Yan, 2011), we need to rethink the capacities of children and the roles they occupy within the family and society (Naftali, 2016). These transformations also resonate with wider debates on changing generational orders and the reconfiguration of childhood and filial duties in non-Western contexts (e.g., Alipio, 2015; Francisco-Menchavez, 2018). Together, these structural and cultural transformations call for a closer examination of how intergenerational relationships are practised and negotiated in everyday life.

Against this backdrop, it becomes crucial to examine how such structural and cultural shifts are lived and negotiated in everyday family relationships. The formation of children’s agency is a process embedded in multiple relationships (Kuczynski, 2003), raising important questions about liushou children’s agency and power in everyday family life. Prior research has primarily examined their well-being or education, often portraying them as passive recipients of adult decisions (Chen, 2024; Ye and Pan, 2011). An exception is Murphy (2020), who centres liushou children’s voices and demonstrates how they can strategically leverage their parents’ emotions to achieve particular goals. This study builds on such insights by adopting a social relational perspective (Kuczynski, 2003; Kuczynski and Parkin, 2007), which views agency as relational, dynamic, and negotiated within interdependent systems of intimacy and obligation. From this perspective, power is not simply possessed but continually renegotiated through processes of conflict, compliance, resistance, and care. Such a lens allows us to move beyond deficit-oriented portrayals and to foreground children’s active, though constrained, participation in family life.

This study listens to liushou children’s voices to explore their agency in reshaping family relationships amid parental migration. Drawing on social relational theory (Kuczynski, 2003; Kuczynski and Parkin, 2007) and the concept of intimate power (Xiao, 2016), we analyse how children negotiate authority, reciprocity and care within what we conceptualise as a “power triangle” linking migrant parents, co-resident grandparents and children. In doing so, the paper extends social relational theory through the lens of intimate power and contributes to ongoing efforts to theorise generational relations and child agency in non-Western contexts.

Transformations of filial duty, childhood, and agency in migrant family contexts

Filial piety has long been the cornerstone of Confucian familism, shaping hierarchical parent-child relations grounded in obedience, old-age support, and patrilineal continuity (Santos and Harrell, 2016). Since the onset of economic reforms in the late 1970s, while filial piety remains widely endorsed (Deutsch, 2006), its practice has transformed significantly. In China’s socialist market economy, where merit and competence rather than seniority determine social rewards, power has shifted towards younger, better-educated generations, while the social status of ageing parents has declined (Murphy, 2020; Yan, 2011). Rural elders, though still receiving material support, increasingly shoulder childcare and household responsibilities, including the care of liushou grandchildren (Liu, 2017). These shifts have fostered more reciprocal and emotionally intimate intergenerational relations than those prescribed by traditional filial norms (Gu, 2022; Qi, 2016). Contemporary family life is now characterised less by authority and obedience than by multidirectional flows of care and support between generations (Goh, 2011), challenging the “individualisation” thesis in Chinese family studies and highlighting the active role of each generation in renegotiating filial relationships (Qi, 2016; Yan, 2011).

In post-reform China, families have become increasingly child-centred. The traditional belief in raising children as a form of economic security in old age has shifted towards expectations of emotional connection and attachment (Yan, 2011), echoing cross-cultural findings on the changing value of children from economic utility to emotional fulfilment (Zelizer, 1994). This transformation has been intensified by industrialisation, urbanisation and family restructuring (Gu, 2022). Qualitative studies document the growing centrality and “pricelessness” of children in family life, with children often receiving emotional and educational investment from both parents and grandparents (Fong, 2004; Goh, 2011; Qi, 2016). The “little emperor” phenomenon illustrates how only children may exercise increased agency in interactions with multiple caregivers, reflecting shifting parental values and reconfigured filial expectations, particularly in urban China (Fong, 2004; Naftali, 2016).

Yet such transformations are highly uneven. While urban families increasingly valorise the child as an emotional centre, children in rural and migrant-sending areas continue to face constrained opportunities and vulnerabilities. Numerous studies have documented how liushou children experience disadvantages in education, mental health, caregiving quality and social support (Chen, 2024; Ye and Pan, 2011). These disparities reveal that the so-called “child-centred turn” in Chinese families is deeply stratified: empowerment and vulnerability coexist within the same moral economy of care. It is within this tension that the question of children’s agency becomes most salient.

While liushou children’s vulnerabilities are often described in terms of material deprivation or disrupted care, they are equally embedded in the moral and emotional economies that bind them to their families through expectations of reciprocity. Liushou children benefit from care, support, and educational investment provided by both parents and grandparents (Gu, 2022; Murphy, 2020), yet they are also morally obliged to reciprocate. A culturally ingrained sense of indebtedness to parents for birth and nurturance has long shaped child socialisation in China (Gu, 2022), further intensified by the moral narrative of parental migration as self-sacrifice: migrant parents’ hardships and indignities as low-status urban workers are portrayed as endured for their children’s future (Murphy, 2014). Recognising such sacrifice generates feelings of guilt and reinforces filial duty. Empirical studies show that liushou children reciprocate not materially but through academic diligence and emotional labour (Gu, 2022; Murphy, 2014). The harsher the parents’ dagong (migrant work) experiences, the deeper the children’s awareness of sacrifice, and the stronger their perceived obligation to demonstrate filial piety through obedience and hard work (Murphy, 2020). A good report card is thus regarded as the most meaningful form of repayment to the family, while failure provokes guilt and frustration (Gu, 2022).

While such moral economies of reciprocity are deeply rooted in Chinese cultural frameworks, they are not unique to China. Similar dynamics can be observed in other transnational contexts where parental migration is framed as sacrifice and children are socialised into reciprocity. Studies of Filipino and other migrant families show that parents maintain emotional closeness and encourage diligence by invoking their own hardship (Dreby, 2010), while children act as active agents in sustaining the migration project through their interpretations of reciprocity (Alipio, 2015; Francisco-Menchavez, 2018; Hoang et al., 2015). Alipio’s (2015) research on the ethic of reciprocity among Filipino children reveals how they fulfil filial obligations through the prudent use of remittances and the maintenance of intergenerational ties. Dreby (2010) likewise finds that children assume household duties as a form of moral repayment. Francisco-Menchavez (2018) further shows that children’s emotional and care work, sustained by guilt and love, keeps transnational families functioning: they are often pressured to study hard and complete college as an affective form of reciprocation – a pattern that closely parallels Chinese cases where sacrifice and academic diligence intertwine (Murphy, 2020).

Together, these studies highlight a central paradox: children’s actions within migration-affected families are simultaneously constrained by moral obligations and enabled through affective ties. Their participation in reciprocity through study, care, and emotional labour illustrates that agency arises not from autonomy but through interdependence. Within this ambivalent moral terrain, a relational understanding of children’s agency becomes essential. Agency, defined as “the ability to make sense of the environment, initiate change, and make choices” (Kuczynski, 2003: 9), is best understood as an ongoing, negotiated process within relationships (Sevón, 2015). It may take the form of compliance, resistance, or accommodation to the status quo (Kuczynski and Parkin, 2007; Punch, 2005), continually reshaped by the conditions of migration. Research from Southeast Asia and other non-Western contexts likewise shows that left-behind children, though excluded from major decision-making, influence care arrangements, sustain family ties, and cope with parental absence through affective and moral labour (Hoang et al., 2015; Lam and Yeoh, 2019). At times, their marginal status even affords them limited bargaining power, as parents and elders become more lenient out of guilt (Dreby, 2010). Building on these insights, this study examines how power is constituted and redistributed among children, parents, and grandparents in multigenerational liushou families, focusing on the everyday negotiations through which authority, obligation, and care are enacted and contested.

Theoretical framework

Childhood can be understood as a relational concept connected to parenthood, in which “the social construction of childhood and adulthood involves a process that includes the agency of both children and adults” (Punch, 2005: 169). Rather than viewing children as isolated actors, social relational theory (Kuczynski, 2003; Kuczynski and Parkin, 2007) conceptualises them as embedded in systems of interdependence, where power derives less from individual resources than from what can be mobilised within relationships. This framework offers a dynamic alternative to unidirectional models of socialisation by emphasising interaction, reciprocity and the recursive exercise of agency. Children and adults continually interpret each other’s behaviour, anticipate reactions and adjust strategies through compliance, persuasion, resistance and accommodation.

Within this framework, negotiation is central. Descriptively, it captures the everyday exchanges through which children and adults respond to one another, ranging from persuasion and compromise to subtle forms of resistance (Kuczynski and Parkin, 2007). Analytically, negotiation explains how intergenerational power is continually redistributed while intimacy is preserved. It is not merely a communicative skill but a relational strategy through which power circulates, becomes affectively charged and is moralised in everyday life. Kuczynski and Parkin (2007) suggest that children’s apparent opposition is rarely aimed at defiance but at restoring cooperation, making negotiation a constructive and dialogical form of resistance. Sevón (2015) further argues that parent-child power is co-constructed in daily interaction while remaining embedded within generational asymmetries. Conceptualising children’s agency as both resistance to constraint and the capacity to influence adults reveals the ambivalent space in which agency is simultaneously enabled and limited (Kuczynski, 2003).

In liushou families, negotiation mediates between the distant authority of migrant parents, the caregiving role of grandparents and children’s moral obligations at home. Such interactions are pragmatic and moral, shaped by filial piety, sacrifice and reciprocity (Gu, 2022; Murphy, 2020). To deepen this analysis, the concept of intimate power (Xiao, 2016) highlights how emotional bonds both enable and constrain authority. Building on Chinese family scholarship, Xiao views family politics as a dynamic process in which members pursue voice and respect by accumulating moral capital and maintaining cooperation. Power here is fluid and relational, circulating through acts of care, respect and emotional labour rather than through fixed hierarchies.

Unlike general notions of intimacy, care or even agency, intimate power does not locate power within the autonomous subject but within the relationship itself. Authority and resistance are mediated through intimacy: the capacity to act (agency) is inseparable from the obligation to sustain closeness. Intimate power entails three interlinked dimensions: intimacy as a precondition for influence, affective ties as the medium of negotiation, and voluntary compromise as a means of preserving harmony (Goh, 2011). In Xiao’s formulation, this intertwining of dignity and intimacy produces tension and coordination within families, as members continuously balance cooperation and self-assertion.

Approaching liushou families through the lens of intimate power offers a more nuanced understanding of how migration reshapes intergenerational relations. Migration does not simply change family structures; it reconfigures the moral and emotional logics through which authority, care and obligation are experienced (Guo, 2025). Rather than viewing these relationships in terms of either harmony or conflict, the framework highlights how dependence and autonomy, affection and control coexist in continual negotiation. This perspective enables us to analyse the family as a relational field in which influence circulates through emotions and moral expectations, and where children’s agency is inseparable from their participation in sustaining intimate ties. In this sense, intimate power extends social relational theory by offering an analytical lens for exploring how families remain connected and unequal across distance.

Methodology

This study adopts an ethnographic approach to explore how liushou children in rural Sichuan negotiate power and agency in the absence of their migrant parents. Fieldwork was conducted over a 9-month period between 2021 and 2022, beginning with a pilot study of more than 20 days, followed by over 8 months of sustained ethnographic immersion in Yunye Town, a migrant-exporting community with around 1300 liushou children. The ethnographic orientation was essential for capturing the subtleties of everyday interactions, silences and ambivalences that often elude structured surveys or psychological testing (Madden, 2017). By centring children’s perspectives, the study foregrounds how authority and intimacy are lived and contested in daily practices.

The research involved 26 children aged between 7 and 17 years, including nine boys and 17 girls. All participants were identified as liushou because one or both parents had migrated for work, leaving them in the care of grandparents or other relatives. Recruitment was based on voluntary participation, with both parental and children’s consent obtained in advance. Access was facilitated through local networks, while rapport was established by offering voluntary English tutoring and everyday companionship. The first author’s fluency in the Sichuan dialect, together with family roots in the region, further supported acceptance and trust within the community. Such trust-building was particularly important given the sensitivity of family conflicts and intergenerational discipline (Nowicka and Ryan, 2015).

Data were generated through a combination of semi-structured interviews and participant observation. A total of 26 interviews were conducted with the children, each lasting between one and 2 hours and addressing themes such as parental migration, discipline, conflict and everyday negotiations. All interviews were carried out in the Sichuan dialect, recorded with consent and subsequently transcribed into Mandarin for analysis. To complement the interviews, the first author immersed himself in the daily lives of five families for periods of one to one-and-a-half months with each household, joining meals, observing homework routines and walking children to school. Three types of field notes were recorded: jottings taken in situ, extended field notes written shortly after interactions, and reflexive memos that captured evolving interpretations and methodological dilemmas (Emerson et al., 2011). This multi-method strategy enhanced the depth and contextual richness of the data.

To strengthen the trustworthiness of the findings, interviews and observations were combined to enable triangulation, key themes were revisited with participants to ensure that interpretations resonated with their perspectives, and the researcher remained in the field long enough to observe both routine and extraordinary events. Such prolonged engagement enabled thick description (Madden, 2017), capturing not only what participants said but also how they lived their daily negotiations of power. These strategies collectively enhanced the credibility and dependability of the study (Lincoln and Guba, 1985).

Analysis was conducted iteratively using NVivo software. We began with inductive open coding of transcripts and field notes, identifying preliminary themes around conflict and resistance, autonomy, care and concessions, and intergenerational alliances. These codes were gradually refined through constant comparison across cases, leading to a thematic framework that highlighted the triangular dynamics between children, migrant parents and grandparents. Reflexive memos played a central role by enabling us to interrogate our assumptions and trace how interpretations evolved. The process was cyclical rather than linear, moving between data and theory in ways consistent with qualitative rigour (Braun and Clarke, 2006). Attention was also paid to negative cases and silences, ensuring that the analysis did not overstate consensus but acknowledged contradiction and ambivalence.

The study received ethical approval from UCL. To protect participants’ privacy, pseudonyms were assigned to all children. Reflexivity was integral throughout the fieldwork, allowing for critical assessment of our positionality and assumptions during interviews and observations (Punch, 2002). As the first author conducted the ethnographic fieldwork, particular care was taken when addressing the sensitive topic of family conflict. Rather than intervening in caregiving practices, the first author concentrated on listening to children’s feelings, offering empathetic responses and engaging in discussions, while ensuring confidentiality to safeguard their privacy. This approach fostered mutual trust and facilitated a more ethical research process.

Findings

Conflict and negotiation: Everyday tensions in the power triangle

Conflicts and contradictions are inherent in parent-child relationships, shaping everyday negotiations of power and intimacy (Kuczynski, 2003). In Chinese families, grandparents have long shaped such dynamics, but in liushou households parental migration reconfigures their role from supportive caregivers to primary disciplinarians enforcing parental authority. As a result, conflicts unfold not only within dyadic relations but also within a triangular configuration of migrant parents, co-resident grandparents and children. This “power triangle” is continually negotiated as authority, intimacy and resistance shift across relationships. From children’s perspectives, tensions arise through remote parental surveillance, grandparents’ discipline and disagreements between parents and grandparents. In what follows, we examine each side of this triangle to trace the frictions and negotiations that shape children’s agency.

Children-parents: Remote authority and ambivalent resistance

The everyday encounters between liushou children and their migrant parents were shaped by a distinctive form of discipline that relied on distance and technology. As (Madianou and Miller, 2012; Peng and Wong, 2016) observe, digital technologies often become tools of “polymedia” parenting, extending migrant parents’ supervision into children’s daily lives. In liushou households, phone calls and video chats were not merely channels of intimacy but also instruments of surveillance, used to remind children about homework, monitor their whereabouts and scold perceived misbehaviour. Meng, a 16-year-old girl, described how her father “calls me at the same time every day” but interpreted this not as affection but as control: “If I’m outside playing with friends when he calls, he will criticise me immediately. It feels like he is watching me all the time.” Unlike non-liushou families, where discipline often takes place face-to-face, parental authority here was mainly mediated through technology. Yet, as Landolt and Da (2005) suggest, migration does not result in family disintegration but compels families to renegotiate established codes. Threats sometimes lost their force because children knew punishment might not be enforced until months later. As Qing, aged 12, recounted with laughter: “My father says he will beat me with a stick if I don’t do well in exams, but by the time they come back for Spring Festival, they already forget.” Migration thus weakened the immediacy of parental discipline, forcing both sides to continually redefine its meaning.

However, children did not simply evade parental authority. Instead, they developed tactics of avoidance, concealment and performance to navigate this terrain. Some deliberately withheld information, offering parents only censored versions of their lives. Fan, a 17-year-old boy, admitted: “When they tell me not to stay up late, I just read novels secretly and don’t reply to their messages, so they think I’m asleep.” This resonates with Pantea’s (2011) study of Romanian left-behind children, who used selective disclosure to undermine remote authority, leaving parents with only partial knowledge.

Others feigned compliance to soothe parental anger. Lan, a 10-year-old, explained: “If I hang up when she scolds me, she will keep calling back. If I act cold, she gets angrier. So I pretend to be enthusiastic, then her anger disappears.” Yun, aged 15, echoed this strategy: “I just humour him or keep silent, let him release his anger, and make him feel I am obedient. I’m just going through the motions.” Such examples illustrate that children rarely engaged in direct confrontation. Instead, they relied on strategic compliance, affective performances and silence to manage parental expectations. These strategies highlight the ambivalence of resistance: they enabled children to claim moments of autonomy yet simultaneously reproduced parental authority by acknowledging its legitimacy enough to require management.

Children such as Meng and Qing openly doubted the effectiveness of remote threats, yet their performances of obedience confirmed that parental commands still mattered. In other words, the very need to resist demonstrated the enduring weight of parental authority. Moreover, this authority was never sustained by digital technologies alone. Co-resident grandparents often acted as enforcers, lending material force to otherwise distant commands. Xingxing, a 12-year-old boy, admitted: “Although my mother’s threats are mostly ineffective, if I am too rude or ignore her, my grandparents will scold me or even punish me physically.” This demonstrates how parental authority, even when exercised remotely, was reinforced through intergenerational alliances. As Goh and Kuczynski (2009) argue, children’s relationships with multiple caregivers provide relational resources, but these resources can also reinforce parental control when grandparents act as its guardians.

Hence, the interplay of distance, intimacy and grandparental enforcement produced a particular ambiguity: geographical absence weakened parents’ capacity to discipline, yet emotional bonds and grandparents’ delegated authority sustained their symbolic power. What made this dynamic distinct was the condition of distsant parenting. In non-migrant households, discipline unfolded face-to-face, whereas in liushou families parental authority circulated through phone calls, video screens and the delegated punishment of grandparents. Children’s agency emerged in this interstitial space, not as outright resistance but as a shifting balance of humouring, concealing and negotiating.

Children-grandparents: Discipline, surveillance, and everyday frictions

If parental discipline was diluted by distance, grandparents embodied the opposite dynamic: immediacy and proximity. Living under the same roof, they exercised discipline through direct observation, verbal scolding and, at times, corporal punishment. This immediacy limited children’s room for manoeuvre, making it harder to ignore or evade authority compared with remote parental supervision. As Xuan, a 12-year-old girl, put it, “I don’t dare to talk back to her, and if I do, she will hit me.” Silence, in her case, was not passive submission but a calculated response to avoid escalation. Yet even within this constrained environment, children devised subtle forms of resistance. When ordered to keep “doing homework” after she had already finished, Xuan quietly shifted to writing stories, knowing that “my grandmother cannot read, and she would think I was still studying.”

At times, however, Xuan’s strategies escalated into open defiance. She admitted to arguing with her grandmother when reprimanded, which often led to corporal punishment. Given the power imbalance, her responses were usually limited to giving in or pleading. Such moments exemplify what Kuczynski (2003) calls covert resistance: children in unequal contexts devise hidden or reactive strategies to assert autonomy while maintaining an outward appearance of compliance.

Although face-to-face discipline left children with little room for manoeuvre, digital technologies created new avenues for negotiation. Smartphones, in particular, enabled children to draw absent parents back into household dynamics, sometimes turning parental authority against grandparents. After punishments, Xuan often exaggerated her grandmother’s harshness when speaking to her mother, prompting intervention: “If she beats me, I tell my mother. Sometimes my mother questions her, and then my grandmother stops.” At times, the mere threat was sufficient: “I will tell mother,” she warned, leading her grandmother to replace beating with verbal scolding. Similar strategies have been documented elsewhere. Poeze et al. (2017) show how left-behind children in Ghana mobilised mobile phones to undermine caregivers’ authority by appealing to migrant parents. This case demonstrates that children were acutely aware of the triangular structure: grandparents reinforced parental discipline, yet could themselves be challenged through parental mediation.

Taken together, these stories reveal the distinctive dynamics of liushou families. In many non-liushou families, grandparents are positioned primarily as nurturers, while parents retain the role of disciplinarians (Naftali, 2016). In liushou households, by contrast, grandparents often occupied the front line of daily discipline. Their proximity enhanced their authority, yet it also made them more vulnerable to children’s strategies, whether through silence, selective compliance or by drawing parents into conflicts. Importantly, the effectiveness of these manoeuvres depended on the strength of the intergenerational alliance: when parents and grandparents acted as a united front, children’s leverage diminished (Goh and Kuczynski, 2009). For instance, unlike other children, seven-year-old Yan shared that “telling my mother is useless, she supports my grandmother’s discipline and even encourages her to punish us if necessary.” This highlights how, in some cases, parental endorsement reinforced rather than softened grandparental authority, closing off potential avenues of resistance.

Interestingly, children often described their experiences with ambivalence. Nong, a 15-year-old, reflected, “Parents just tell you their ideas and scold you, but grandparents also wake you up and sit with you. Even if they scold, I feel they are doing it for my good.” Yun, aged 15, added that despite her grandfather’s strictness, “Compared to my mother, I am closer to them.” These accounts underline how power and intimacy are entangled: the same relationships that constrained children also provided emotional support, embodying what Xiao (2016) terms intimate power. Taken together, these accounts show that children’s negotiations with grandparents were never isolated but embedded in broader intergenerational configurations, which they actively observed, a theme explored in the next section.

Parents-grandparents: Alliances, divergences, and children’s observations

Although the previous sections focused on children’s direct negotiations with parents and grandparents, here we turn to how children observe and interpret the relationship between their parents and grandparents. From their vantage point, this intergenerational relationship was far from stable; it oscillated between cooperation and conflict, and children not only witnessed but also actively drew upon these dynamics in shaping their own strategies of survival and resistance. As Melander et al. (2020) note, such “care triangles” are historically and relationally constructed through trust and reciprocity, yet children’s accounts here reveal how complex and evolving these alliances can be in everyday practice.

For many children, the most immediate perception was that parents and grandparents often acted in alliance. Migrant parents, despite their distance, sometimes reinforced their authority through grandparents, issuing instructions that were then executed in the household. Several children remarked that disobeying grandparents felt tantamount to defying parents, since the latter could intervene by phone and validate the discipline. This triangulated enforcement often made children wary of directly confronting either side. At other times, however, they quickly recognised divergences. Yun, a 15-year-old girl, offered a vivid example: because her grandfather strictly forbade mobile phone use, she and her grandmother secretly pooled money from their “private savings” to buy a new phone. For over a year, her grandfather remained unaware, since he rarely entered her room. Yun used this phone to communicate with her divorced mother, despite her grandparents’ disapproval. She carefully waited until her grandfather was out of sight before video-calling, while her grandmother, though displeased, tolerated these conversations because they allowed Yun to share everyday details with her mother. Such a story highlights how children recognised cracks in intergenerational authority and sometimes built small alliances with one caregiver against another. This resonates with Goh and Kuczynski’s (2009) observation that ambiguity and contradiction are inherent in intergenerational relations, which continually adapt and reconfigure over time.

Children also mobilised differences between generations. Qing, a 12-year-old girl, was forbidden by her grandmother to use a mobile phone, which she believed caused myopia and poor academic performance. Her mother, however, regarded it as essential for maintaining contact. Qing explained: “If my grandmother doesn’t let me use my mobile phone, I will ask my mother to negotiate, and most of the time my grandmother will agree.” Guang, a 16-year-old, noticed a broader generational contrast: while her parents and grandparents frequently quarrelled with her, neither her father nor her mother ever physically punished her, even during family reunions. When she asked her father why, he replied that corporal punishment was “not the role of my generation.” For Guang, conflicts with parents often ended with coaxing rather than punishment. These accounts underscore how children interpreted generational differences as a resource in negotiating everyday authority.

Sometimes even conflicts between adults could be turned into opportunities. Jian, a 13-year-old girl, explained: “When my grandmother and grandfather quarrel, I call my father. He will persuade them to stop, and he sometimes may reward me and I get what I want.” Children thus learned to mobilise adult disagreements as resources, reframing intergenerational frictions as opportunities for negotiation. In doing so, they embodied an active role in constructing and critiquing intergenerational relations rather than passively reproducing them (Bühler-Niederberger et al., 2023).

Underlying these narratives is a key feature of liushou families: the coexistence of reinforcement and fragility in intergenerational authority. On the one hand, parental authority could be strengthened when grandparents acted as enforcers. On the other hand, divergences between generations opened cracks through which children could exercise agency. As Peng and Wong (2016) note, substitute caregivers often lack the institutional and moral recognition required to fully discipline left-behind children. Children in this study were acutely aware of this gap, and they used it to negotiate, complain or strategically mobilise their parents against grandparents. By observing, interpreting and acting upon these shifting relationships, children revealed the “power triangle” not as an abstract structure but as a lived, contested and constantly reconfigured reality.

In sum, children’s observations of the parent–grandparent relationship reveal both the constraints and possibilities of intergenerational authority in liushou families. Alliances between parents and grandparents could intensify control, but generational divergences created spaces for children’s agency. Far from being passive recipients of discipline, children learned to read these tensions, mobilise allies and strategically position themselves. Yet their manoeuvres were not only about conflict and resistance. As the next section will show, children also engaged in retreat, accommodation and care, often motivated by guilt towards parents’ sacrifices or concern for grandparents’ frailty. These practices complicate a binary view of resistance and compliance, highlighting instead how power in intimate relations is continually reshaped by both tension and tenderness.

Retreat, intimate power, and children’s agency

Although earlier sections highlighted conflict and negotiation, children’s strategies were not exhausted by resistance alone. Many deliberately retreated or acquiesced in ways that preserved intimacy and sustained intergenerational bonds. Here the notion of intimate power (Xiao, 2016) is analytically useful: power in liushou families was mediated not only by coercion but also by love, sacrifice and mutual dependence. Concessions thus became relational practices through which children sustained fragile ties and negotiated their place within the family. This reframing extends the concept of agency beyond opposition, foregrounding its affective and moral dimensions.

Retreat towards migrant parents often stemmed from a sense of indebtedness. Children framed migration as sacrifice, with parents enduring exploitative work so they could “have a better life.” This produced what Gu (2022) terms children’s emotional labour: managing guilt and obligation to meet culturally sanctioned “feeling rules” of filial piety. Fan, a 17-year-old boy, explained: “My mother asks me to make a video call to her every week. Although I don’t like it, I try my best to comply because I have to fulfil my filial duty.” Guang, aged 16, similarly reflected: “When you’re a child, you might choose to argue, but now I won’t, because I think we’re all family, and they work hard outside.” In these accounts, obedience was not mere compliance but a moral response to parental sacrifice.

This logic resonates with traditions in which filial piety (xiao) has long been central to child socialisation (Qi, 2015). Today, as Gu (2022) shows, filiality is increasingly embedded in intergenerational contracts that balance parental investment with children’s emotional reciprocation in liushou families. Concessions can thus be read as forms of “paying back” parents through deference, even when commands felt intrusive. Distance may weaken intimacy, yet it simultaneously magnified sacrifice, strengthening children’s willingness to retreat.

Children often reframed compliance as repayment of moral debts. Ke, a 12-year-old, expressed tearfully: “My parents experience a hard life in the factory. Their sacrifices are all for us, so we should be filial to them in the future.” Such accounts highlight acquiescence as a moralised practice of gratitude, aligning with Gu’s (2022) point that children’s emotional labour normalises feelings of indebtedness towards sacrifice. Indeed, this labour of emotion was culturally cultivated. Chinese children are taught from an early age that they owe their wellbeing to parental efforts (Murphy, 2020). In liushou families, migration amplified this script: parents’ factory or construction work made sacrifice visible. Hence, even when parental phone calls felt intrusive, children such as Fan or Guang described their compliance not simply as obedience but as repayment.

If filial piety and sacrifice framed retreat towards parents, concessions to grandparents were often grounded in intimacy, frailty and care. Unlike distant parents, whose authority rested on sacrifice, grandparents’ authority derived from daily companionship and the vulnerabilities of old age. Children often tolerated discipline not as obedience but as compassion for ageing caregivers. For example, Fan reflected on his changing attitude: “When I was younger, my grandmother and I argued a lot. Now that I’ve grown up and become more sensible, I tolerate what I can. She has taken care of me for so many years, and now she is so old.” Similarly, Ning, aged 15, worried that quarrels might harm her grandmother’s heart condition: “I now avoid arguing with her. She is in her eighties and has coronary heart disease. I even advise my mother to avoid arguing with her as much as possible.” For Ning, restraint was motivated less by filial piety than by recognition of bodily vulnerability.

These accounts resonate with Xiao’s (2016) notion of intimate power: authority mediated by affective ties and sustained through negotiation and compromise. Grandparents’ discipline was often inseparable from care – waking children for school, overseeing homework, preparing meals. For many children, grandparents’ strictness was intertwined with support, as in Yun’s case mentioned earlier, producing a disciplinary relationship that was both caring and constraining. Hence, by tolerating discipline and avoiding confrontation, children enacted what might be called compassionate retreat. As Kuczynski (2003) suggests, intimate relations entail mutual dependence: grandparents relied on children’s compliance to sustain authority, while children recognised vulnerability and chose to concede. Retreat here was not passive submission but a relational choice that preserved fragile solidarity within the household.

Notably, children’s retreats were not unidirectional; parents also yielded in moments of guilt or affection, which children could anticipate and exploit. Shu, aged 11, distinguished between her grandmother’s frugality and her mother’s indulgence: “My grandmother does not allow spending money to take me to the amusement park, but my mother is different because she rarely comes back. So when she comes back, she hardly refuses any of my requests. When she comes back for the Spring Festival, I want her to take me to the amusement park and to eat hot pot.” Here, maternal retreat is rooted in absence, as the mother compensates for long separations by acceding to requests. Shu, in turn, anticipates this retreat and strategically times her demands. Thus, these children are not only making concessions; they also actively perceive parents’ vulnerabilities, whether guilt, love, or fear of conflict, and sometimes mobilise them for their own ends.

In conclusion, the concessions children make in liushou families demonstrate that retreat is not reducible to passive compliance. Rather, it is a relational and ambivalent practice shaped by distinct affective logics: guilt and sacrifice towards parents, compassion and frailty towards grandparents, and even tactical manipulation of parental guilt in everyday life. These practices reveal how authority in intimate relations is sustained not only by coercion but also by love, vulnerability, and reciprocity. As Xiao (2016) argues, intimate power emerges when authority and affection become mutually constitutive, and children’s experiences vividly illustrate this dynamic. By recognising retreat as a form of agency, we move beyond dichotomies of resistance versus compliance. Children’s ambivalent practices, resisting in some moments and conceding in others, show how intergenerational obligations are continually negotiated within the “power triangle.” In this context, retreat is not the absence of action but an active labour of sustaining fragile solidarities. It is precisely in this oscillation between conflict and concession that the everyday politics of liushou families unfold.

The power triangle beyond separation: Migration, intimacy, and agency

This study has examined the shifting dynamics of authority, intimacy, and resistance in Chinese liushou families, drawing on children’s perspectives to conceptualise a new configuration of intergenerational power. By focusing on how children interpret and navigate everyday interactions with migrant parents and co-resident grandparents, the analysis highlights the emergence of a “power triangle,” in which power circulates between absent parents, caregiving grandparents, and children themselves. This triangular arrangement illustrates how migration unsettles established hierarchies while at the same time generating new forms of relational negotiation (Landolt and Da, 2005).

A central insight of this study is that children’s agency in liushou households is marked by ambivalence. Children did not simply comply with or reject authority; rather, they shifted between resistance and concession depending on circumstance and relational context. Remote supervision by parents, often mediated through phones and video calls, provoked subtle forms of resistance such as concealment, humouring, or selective disclosure. These practices enabled children to carve out spaces of autonomy, yet they also reproduced parental authority, since their very enactment acknowledged that commands continued to matter. Similarly, silence or apparent acquiescence towards grandparents’ discipline was rarely passive. It often reflected calculated retreats grounded in compassion for ageing caregivers or in a desire to preserve household harmony. These practices resonate with Kuczynski’s (2003) argument, which emphasises the creativity of subordinate actors in highly asymmetrical relationships.

Seen from this perspective, agency takes relational forms that complicate dichotomies of obedience and resistance. Concessions to parents were frequently narrated as moral repayments for parental sacrifices, echoing Gu’s (2022) observations that children in migrant households often manage guilt and obligation as forms of emotional labour. Concessions to grandparents, by contrast, were often grounded in daily companionship and recognition of frailty. Both forms of retreat highlight that acquiescence can be a deliberate and moralised practice rather than mere compliance. By foregrounding this ambivalence, the study moves beyond portrayals of left-behind children as either passive victims or heroic resisters (Spyrou et al., 2018) and instead reveals them as relational actors who continually balance autonomy, obligation, and intimacy in the everyday politics of family life.

The findings also refine our theoretical understanding of power in family relations. Authority in liushou families was not exercised solely through coercion but continually reshaped by intimacy, vulnerability, and reciprocity. The notion of intimate power (Xiao, 2016) helps to capture this dynamic. Children often tolerated or even valued grandparents’ strictness because it was inseparable from acts of care, such as preparing meals or overseeing homework. Likewise, parents’ long-distance discipline was interpreted through the lens of sacrifice, since migration itself was framed by both adults and children as a costly act of devotion. In these ways, authority was sustained less by unilateral command than by the affective ties binding generations together.

At the same time, the analysis demonstrates that negotiation is not merely a descriptive label for everyday disagreements but an analytical lens for understanding how relational asymmetries are managed. Negotiation occurred when children balanced resistance and retreat, when grandparents yielded to avoid quarrels with migrant parents, or when parents retreated out of guilt. Such practices illustrate that power is not a static attribute of one generation but a process of adjustment and compromise embedded in interdependence (Kuczynski and Parkin, 2007). liushou families thus extend insights from social relational theory by showing how interdependence itself is reconfigured under conditions of migration.

Placing these findings in a wider perspective shows both continuity and divergence with other family contexts. Many of the practices observed, such as appeals to one caregiver against another or tactical compliance, can be found in non-migrant rural families, where generational hierarchies and conflicts also persist (Goh and Kuczynski, 2009). What distinguishes liushou households, however, is the structural condition of prolonged parental absence. This absence makes remote supervision central to parenting, elevates grandparents from nurturers to disciplinarians, and gives children unique leverage through technologies of communication (Guo, 2025). The result is a distinctive triangular configuration in which authority circulates unevenly across generations, continually rebalanced through alliances and conflicts.

Taken together, the findings demonstrate that power in liushou families is best understood as situational, fluid, and relational. Children do not stand outside intergenerational hierarchies but participate actively in their reconfiguration. They resist, conceal, and negotiate, yet they also retreat and concede in ways that preserve fragile solidarities. These practices are inseparable from cultural narratives of sacrifice, filial obligation, and intergenerational care (Gu, 2022; Murphy, 2020), but they also highlight children’s ability to interpret, adapt, and act upon these norms in everyday life.

In conclusion, this study contributes to ongoing debates in the sociology of childhood and family sociology by showing how migration reconstitutes power relations within multigenerational households. The concept of the “power triangle” moves beyond the binary of parental absence and presence, opening up a more layered understanding of intergenerational authority in contexts of mobility. More broadly, the analysis illustrates how authority, intimacy, and agency are mutually constitutive, continually reshaped by both tension and tenderness. By centring children’s perspectives, the study reveals the everyday politics through which family obligations are sustained, reinterpreted, and at times contested. These insights extend beyond the Chinese case, offering a lens for examining how migration and mobility worldwide reconfigure not only who cares for children but also how power, love, and responsibility are distributed across generations.