Perching birds or scattered streams: a study of how trust affects civic engagement among university students in contemporary China

This article explores how trust affects the nature and characteristics of civic engagement among university students in contemporary China. We draw upon empirical evidence from in-depth individual interviews involving 68 students born between 1993 and 1999. Four different types of civic activities, including associational membership with the Chinese Communist Party as well as self-organised town-fellowship associations, volunteering and digital activism, are analysed in relation to bridging, bonding and/or obligatory relationships. Generalised trust amongst university students seems to be low, but they do exhibit unquestioned trust in the State. The organisations with which they associate either are top-down through the Communist Party, which is ideologically narrow and socially distant among the members, or promote geographical origin-based bonding rather than bridging relationships. Our findings add new depth to our understanding of the relationships between the dominant modalities of trust and the typical form of civic participation in Chinese universities. We also highlight the effects of increasing meritocratic competition and stress on performativity on the civic attitudes and behaviour of students. We argue that the hybrid of neoliberal individualism and Confucianism seems to be a perfect match to reconcile the absence of civic traditions and to validate civic consciousness among university youth.


Introduction
There have been intensive episodes of youth activism across the globe in recent years, such as the youth climate action and the Hong Kong youth protests in 2019. Despite some predictions that the Chinese youth will next be involved in the global trend of activism and mobilisation (Wang and Hernández 2020), our understanding of Chinese young people's civic activities has yet to be informed by empirical evidence. Meanwhile, there have been increasing numbers of youth enrolled in universities during the last two decades, with gross enrolment ratios of 48.1% in 2018 (MOE 2019a). The impact of higher education expansion on civic activities amongst university students is still under-researched. What are the patterns of civic activities among contemporary university youth? How do higher education experiences shape their civic engagement, particularly in relation to social and political trust? This study will make a start in using first-hand empirical data to explore how trust affects the nature of civic engagement of university students in China.
There has been a long tradition of researching trust and civic engagement to make sense of the civic nature of a society (Banfield 1958;Fukuyama 1995;Putnam 2000). The conceptualisations of civic engagement are highly contested; however, existing scholarship has identified expressive or competitive political activities, associational membership, activism and volunteering as key measures of civic engagement (Ehrlich 2000;Putnam 2000;Campbell 2006). Trust, broadly speaking, has two forms: political and social trust (Tao et al. 2014;Huhe 2014). The former captures the trust in vertical or hierarchal relationships with the State or a particular institution (Williams 2020), whilst the latter refers to interpersonal relationships at the horizontal level, which are most commonly divided between particularised and generalised trust (Delhey et al. 2011;Uslaner and Conley 2003). Particularised trust connects 'in-groups' such as families, extended kinship and face-to-face communities in which people know each other and interact closely, and where social sanctions are strong (Fukuyama 1995;Portes 1998;Delhey et al. 2011;Portes and Vickstrom 2011). Generalised trust is imbedded in interactions with unfamiliar people (Newton 2007;Nannestad 2008), or the 'out-group', from which a sense of engagement, tolerance, prosperity and democratic relationships are developed (Delhey et al. 2011;Putnam 1995Putnam , 2000. The competing natures of particularised and generalised trust have been the central focus of research on the relationships between trust and civic engagement. Barrington Moore's (1966) seminal book adopts a comparative perspective and analyses the presence and absence of generalised trust in determining different civic traditions and political trajectories at the turn of the twentieth century. Edward Banfield (1958) pioneered the use of interdisciplinary approaches to identify the social, moral, political and cultural characteristics of particularised trust in a poor southern Italian village, which were attributed to the underdevelopment of a meaningful civil society and collaborative networks essential to an advanced economy. Fukuyama (1995) further explores the cultural roots of particularised and generalised trust across different contexts and argues that the characteristics of social trust in a society explain different levels of associational civic organisations, economic development and effectiveness of political institutions.
Civic engagement, in turn, influences trust. Bonding and bridging social capital are used to differentiate the exclusiveness and inclusiveness of civic activities and their different implications for the level and radii of social trust (Delhey et al. 2011;Putnam 1995Putnam , 2000. The former refers to a narrow radius of socioeconomic, cultural and psychological networks for solidarity and reciprocity among exclusive and homogenous membership, which promotes particularised trust and lowers the level of generalised social trust (Putnam 1995(Putnam , 2000. By contrast, bridging social capital provides crucial social links and space among those from diverse social, political, cultural and ethnic backgrounds (Putnam 2000: 22), which helps to develop generalised trust.
Furthermore, many scholars take advantage of the availability of cross-national survey data such as the World Values Survey to develop systematic analysis of the interplay between trust, bonding and bridging civic activities in contemporary societies (Nannestad 2008;Delhey et al. 2011Delhey et al. , 2014Delhey and Welzel 2012;Frederiksen et al. 2016). Although the dominant findings from cross-country or single-country studies do not show a consistent relationship between the levels and radii of social trust and the levels of civic engagement, there are still many valuable insights into the variations at the individual and contextual levels. Some context-level research finds that a range of civic activities, such as civic group membership, charitable donations and volunteering, boost bridging relationships and are positively correlated with the level and radius of generalised trust (Delhey et al. 2011;Uslaner and Brown 2005). By contrast, the strong in-group associations with a heterogeneous membership base are more likely to strengthen particularised trust. For instance, Uslaner and Conley's (2003) work on Chinese immigrant minorities in the USA suggests that the members of ethnic Chinese civic associations are less likely to participate in society-wide civic activities or collective action than, for instance, ethnic Chinese US citizens in general.
Other research highlights individual-level variables, including socioeconomic origin, education, gender, geographical characteristics, psychological and emotional factors, and personal motivations and goals (Tao et al. 2014;Barrett and Brunton-Smith 2014;Frederiksen et al. 2016;Wu and Shi 2019). For instance, a high level of education is linked to a wide range of bridging civic activities, from community-level associational membership, to volunteering and donations, to environmental activism and feminist movements, to political participation (Jayakumar 2008;Doyle and Skinner 2017), which in turn promotes generalised trust (Wu and Shi 2019). Relatedly, Barrett and Brunton-Smith (2014) find some evidence that people of higher socioeconomic status tend to have a higher level of civic engagement compared to people from lower socioeconomic backgrounds in some European countries. This might be mediated by the fact that people from higher socioeconomic origins are likely to have a higher level of education and civic knowledge and experiences. However, Silva's (2019) recent study of the American working class contradicts the powerful effects of education on political and civic engagement. Instead, she provides a poignant analysis of how a lack of political trust of mainstream politicians and institutions inspired different forms of grassroots civic engagement among the alienated working class, which promotes bridging relationships.
These studies make significant contributions to our knowledge of interdependent relationships between civic engagement and trust. However, these studies fall short when explaining the Chinese context. Context-level studies on trust tend to use all-embracing Confucianism as the key indicator to explain the Chinese case. Confucianism is used to explain the high level of particularised in-group trust but the low level of generalised trust in East Asian countries such as China and South Korea (Delhey et al. 2011). However, this Confucianism account does not apply to Taiwan and Japan, where the level of generalised trust is relatively high (Fukuyama 1995;Yamagishi et al. 1998).
Furthermore, studies on the civic nature of Chinese society suggest that there is an absence of a fully developed civil society with active and efficient associations (Moore 1966;Rankin 1993;Pye 1999). Following this tradition, recent studies further focus on expressive political activities, such as labour movement activism, as empirical cases to illustrate the cultural and institutional obstacles to developing a meaningful civil society in contemporary China (Fu 2017). However, these studies do not often allow for an investigation of 'ordinary' citizens and their civic engagement. Moreover, they do not pay sufficient attention to participatory political activities, civic activities or volunteering associations, nor do they specify the demographic characteristics of active citizens in different patterns of civic engagement. Most importantly, there is an evident gap in the use of trust to explain different patterns of civic engagement. Therefore, this study will focus on the patterns of civic engagement among a selected population-university students-and examine how these patterns of youth engagement are shaped by trust in contemporary China.

Trust in the Chinese Context
Existing scholarship on trust in China takes Confucianism as the central theoretical point of departure (Fukuyama 1995;Pye 1999;Delhey et al. 2011;Dickson et al. 2017). Although there is no definite interpretation, Confucianism can be conceptualised by a set of cultural norms, values and codes, including filial piety, familism, belief in meritocracy, self-cultivation, acceptance of hierarchy, respect for teachers, respect for authority, loyalty to the State, the responsibility of the family and social harmony (Rankin 1993;Fukuyama 1995;Pye 1999;Dickson et al. 2017). Prior studies consistently highlight a unique pattern of obligatory relationships with individual families and the State, which are absent from Western literature's emphasis on bridging and bonding relationships (Fukuyama 1995;Huhe 2014). For instance, quantitative studies drawing from large-scale national datasets or international comparative survey data highlight a Confucian model of obligatory trust (Delhey et al. 2011;Dickson et al. 2017). These studies show a high level of trust in individual families and networks (Delhey et al. 2011(Delhey et al. , 2014Huhe 2014), as well as a high level of institutional trust, which embodies loyalty to and support of the State (Dickson et al. 2017), but a low level of generalised trust (Tang and Huhe 2014).
The Confucian core value of familism shapes in-group obligatory relationships (Fukuyama 1995;Pye 1999). Trust and loyalty among family members are unquestioned and sometimes irrational, although they dictate patterns of people's behaviour and choices (Fukuyama 1995;Pye 1999). For instance, previous research has shown a link between the familist obligation and consumer culture, such that people felt obligated to support a business owned by family members or extended kinship rather than follow a rational choice (Fukuyama 1995). The high level of particularised trust can also be extended to quasi kinship networks, such as membership in the same village (Huhe 2014). By contrast, attitudes of mistrust towards outsiders or strangers were found to be responsible for the limited scale of business expansion by private entrepreneurs in Taiwan and Hong Kong during the 1960s (Fukuyama 1995).
Furthermore, Confucian cultural codes, such as respect for authority, loyalty to the State and social harmony, seem to have been reliable sources for the Chinese Communist Party to reinvent Party lines, thus continuously holding public trust during the market reform and recent Xi regime (Wu and Wilkes 2018;Dickson et al. 2017). However, there has been a consistently low level of political trust in local institutions, local governments and civil servants, partly because of deeply rooted cultural distrust in local officials (Pye 1999;Wu and Wilkes 2018) and partly because of the tightly controlled and censored media, which has targeted low-level officials as scapegoats for social problems and corruption (Tang and Huhe 2014).
Perhaps the most unique characteristic of trust in the contemporary Chinese context is the role of Chinese Communist Party membership in the development of generalised trust.
Drawing upon the data on rural villagers from the China General Social Surveys, Huhe (2014) finds that Chinese Communist Party members tend to show a higher level of generalised trust towards strangers than non-CCP members among rural residents. He further argues that the socialisation skills acquired to be a Party member and the Party's public-serving ideology might contribute to developing a high level of generalised trust (Huhe 2014).
These studies have enriched our knowledge of trust patterns in China. However, they are subject to a number of limitations. Quantitative studies that draw from national survey data do not permit an in-depth investigation of civic activities and networks. For instance, Huhe's research is innovative in contextualising generalised and particularised trust in rural villages in China by measuring the trustworthiness of 'relatives and close friends, neighbours, nonneighbours, villagers with the same surnames, villagers with different surnames and strangers' from the China General Social Surveys (Huhe 2014: 587). However, these datasets do not include information on civil activities such as volunteering. Therefore, they do not allow for an in-depth understanding of how trustworthiness is translated into bonding or bridging relationships. Qualitative research such as that conducted by Pye (1999) and Rankin (1993), which relies on an interpretation of Confucianism, does not seem to utilise empirical evidence or indepth methods to interrogate the relationships between trust and civic engagement in contemporary society. Furthermore, research on the role of Chinese Communist Party membership does not permit a distinction between obligatory and bridging relationships in promoting generalised trust. Moreover, it does not extend to explain the extent to which the Communist Party socialisations facilitate the development of bonding relationships among CCP members or bridging relationships between CCP members and non-CCP members. Therefore, this study will focus on contemporary university youth as a case to explore the patterns of civic engagement and bonding, bridging or obligatory relationships emerging from these activities, as well as how the level of trust from these relationships affects civic engagement.

Higher education, trust and civic engagement
The choice of university students as the primary focus of this study merits some explanation. First, research from other contexts shows higher education plays an important role in shaping civic engagement (Jayakumar 2008;Doyle and Skinner 2017). Formal learning curricula, knowledge production and fields of study not only have tremendous pedagogical effects on enhancing students' civic skills but also affect students' social and civic attitudes. For instance, studies from the USA and some European countries show the effect of classroom teaching and learning on students' social attitudes (Mariani and Hewitt 2008;Elchardus and Spruyt 2009;Gross and Fosse 2012). Stubager's (2008) study, which analyses Danish voters aged between 18 and 75 in terms of different types of education received, finds that the strongest effects on generalised trust come from higher education in medium-and long-term programmes, particularly in social fields of study. Furthermore, university students use their knowledge to advocate for social justice, such as the decolonization of the HE curriculum, de-academised knowledge (Cole and Heinecke 2018: 11) and the #RhodesMustFall campaign in South Africa and the UK (Konik and Konik 2018).
Second, prior scholarship often links higher education with a high level of generalised trust, although the results are dependent on the wider social contexts (Frederiksen et al. 2016). Drawing from cross-national datasets such as the World Values Survey, some research provides empirical evidence on the positive correlations between increasing cognitive capacities associated with higher education knowledge and experiences and high levels of generalised trust, particularly in Western contexts (Yamagishi 2001;Delhey et al. 2011). Further studies argue that the impact of higher education on generalised trust is mediated by wider sociopolitical circumstances, such as levels of corruption and political risks (Frederiksen et al. 2016;Wu and Shi 2019). Frederiksen et al.' (2016) study highlights that the level of education is negatively correlated to the level of generalised trust among high-corruption countries from the WVS data between 2005 and 2009. Similarly, Wu and Shi (2019) argue that people with more education are more likely to adopt risk aversion strategies in high-risk societies, which lowers generalised trust.
However, these perspectives on the role of higher education in shaping trust and civic engagement have yet to be examined systematically in the Chinese context, perhaps with the exception of a recent study on post-Mao education development by Vickers and Zeng (2017). They highlight that the top-down patriotic narratives throughout the education system contribute to a unique pattern of individualised and divided citizens with unquestioned trust in the State (Vickers and Zeng 2017). Relatedly, we know that civic engagement among university students is primarily politically oriented (Li 2009;Zhang 2016;Wang 2016). For instance, Li's (2009) empirical work highlights politically active and involved university students but Zhang's (2016) survey of around 200 university students in Beijing suggests that most students are only passive participants in political activities, responding mainly to pressures experienced from messages emanating from the curriculum on political education. This idea of 'passive citizens' is further reflected in Wang's (2016) study on political attitudes in China. His statistical analysis suggests that political participation in universities does not seem to make a significant impact on graduates' political attitudes compared to those of non-graduates.
We also know that university experiences provide unique socialisation opportunities for people of diverse social and geographical origins (Liu 2016). However, much less is known about whether political activities promote bridging or bonding or obligatory relationships. We also know very little about civic activities in which university students are involved and about how university experiences shape their civic engagement, particularly in relation to social and political trust. In light of this gap, this research asks a number of questions: (1) What types of civic activities are young people involved in at university? (2) How do they perceive interpersonal (bonding, bridging and/or obligatory) relationships during these activities? (3) How does the level of trust emerging from these relationships affect their commitment to civic engagement?

Data, methods and research design
The dataset for this research consists of in-depth, semi-structured interviews with 68 university students from birth cohorts between 1993 and 1999. In choosing these cohorts, we aimed to recruit respondents who were still in higher education institutions at the start of this project in April 2017. In the Chinese language, the concept of trust has many translations in different linguistic settings. The most commonly used term is xinren (信任); however, there are many other synonyms, such as xiangxin (相信), xinlai (信赖), kexin (可信), jianxin (坚信) or kekao (可 靠), kaopu (靠谱), which can be used in a variety of formal and colloquial settings. Because trust is an abstract concept and is invisible, in-depth interviews enable us to uncover the levels of trust deeply buried in the complex socialisation and activities experienced by these young people. The interviewees were selected from a variety of social backgrounds as well as from different geographical origins. We approached students in canteens, sports centres, libraries and student-organised societies on campus, and we also recruited respondents using social media websites and applications such as WeChat and QQ. Table 1 summarises the socioeconomic and demographic characteristics of the 68 interviewees, university students and graduates, from the birth cohorts between 1993 and 1999. This table also includes the pathways of higher education-that is, the types of universities and fields of study of the interviewees. Table 1 shows that 16 respondents in our sample come from key universities, whilst 52 come from non-key institutions. Key universities refer to worldclass universities and prestigious national universities in the 211 project (Liu 2018). The proportion of interviewees from different types of universities merits some explanation. The Chinese higher education system is hierarchal in nature, with more than four times as many non-key universities as elite and key universities (Liu 2015). Furthermore, the data obtained from the Ministry of Education show that undergraduates registered in the 211 universities accounted for only 10.25% of the total undergraduate population in 2018 (MOE 2019b). Therefore, we chose to recruit more respondents from non-key universities than from key institutions, although this research does not aim to provide a generalisable picture of students from different types of institutions. Regarding the fields of study, we primarily distinguish the STEM subjects from Arts, Humanities and Social Sciences, with 36 and 32 respondents from these respective pathways.
The interviews were semi-structured and lasted approximately two to two-and-a-half hours. All interviews were conducted in Mandarin Chinese, audio-recorded with the interviewees' consent, transcribed in Chinese, translated into English and analysed in English. The students' identities and institutions were anonymised, and pseudonyms were coded instead. Coding and mapping techniques were used to analyse the interview data (Fig. 1). Open coding was used initially to identify the patterns of non-academic or extra-curricular activities and experiences, which are coded into organised and expressive political engagement and organised and self-organised civic engagement. We then identified interpersonal bonding, bridging or obligatory relationships with local communities, institutions and the State by synthesising these patterns. Drawing upon the synthesised codes, we mapped the connections between interpersonal and institutional trust and the patterns of individuals' civic engagement. Figure 1 is introduced as a coding schema-a degree of civic engagement determined by the consistency of individuals' commitment to civic activities and meaningful bonding, bridging and/or obligatory relationships emerging from these activities. The level of individuals' trust manifested through these interpersonal relationships is the key to understanding the different degrees of civic engagement.

Findings
Town-fellowship associations, particularised trust and bonding relationships We identify a high level of engagement in a particular type of associational activities and events relating to town-fellowship associations (N = 63). These are based on geographical origin (village, town, county, city or province). Differing from the traditional fraternity clubs in US universities, these associations do not often have an office or a building on campus; instead, they are represented by rather 'fluid' but organised groups of students by enrolment year, field of study or residency halls. These representatives organise social events as well as virtual groups on WeChat. A total of 63 respondents confirmed that they participated in townfellowship societies, and 41 of the 63 respondents were regulars who were committed to building up their networks and helping their town fellows. Zhikai Liu, a 24-year-old postgraduate from a rural area in Hubei, belonged to a Hubei town-fellowship association on campus. Zhikai was an active volunteer for his town-fellowship association. Feeling homesick and lonely away from his hometown when he first came to study in Beijing, Zhikai joined the town-fellowship association in his freshman year. The generous support and help from the senior town-fellows at the same university made him feel committed to the activities organised by the association: I never left Hubei before. It was scary to live in this enormous city. I had no one to turn to. But I was lucky one day when I overheard someone speaking my dialect in the canteen. It was like finding my family again. Our town-fellows helped me get settled in the dorm and gave me a lot of advice on studying and living in Beijing. We had a great sense of solidarity. I like to pass the torch of solidarity by helping other young townfellows.

Higher Education
When asked whether there were other types of student organisations available to support him at that time, Zhikai said: 'There were many clubs and associations on campus, but I felt very lost and did not belong to any particular group. My town-fellow's dialect reminded me of my hometown. He was trustworthy, like a family'. Zhikai made a priority of organising activities and took the initiative to set up a social media account-specifically a WeChat group-which helped recruit newcomers to the university and promote events related to job hunting and graduate school opportunities. His contributions to his town-fellowship association 'made him proud', and the time and commitment he gave to these activities made his 'university life more meaningful'. Zhikai's narrative revealed a pattern of networks connecting students from the same town/ dialect with shared interests and solidarity, which resembled the quasi-patrilineal socialisation that prioritises social relationships based on geographical origin. This type of socialisation strengthens bonding relationships between individuals and communities from the same areas, fostering particularised trust. Compared to other types of student organisations, the townfellowship associations seem to represent trustworthiness for students like Zhikai. These students still rely on patrilineal and patrilocal cultural identifications, such as a dialect, to verify trustworthiness, thus consciously filtering in-and out-groups during university socialisation.

CCP membership and mixed obligatory relationships
Another type of consistent civic activity is engaging with the Chinese Communist Party branches on campus. For more than half of the respondents (N = 43), organised political activities constituted the most normal pattern of civic engagement, ranging from applying for Communist Party membership to following the Party's probationary performance requirements to being actively involved in local Party conferences and events. The students explained that these activities were primarily concerned with 'studying' or 'discussing' new policy proposals or new speeches by the statesmen. They confirmed their absolute loyalty to the Party State and accepted all the Party's 'policies and strategies of national development'. They further stated that 'the Party is the best representative of public interests'. The most common pattern among these narratives is the continual 'identity switching' between 'Wo' and 'Guo Jia' ('I' and 'Nation Family'), which suggests a close affinity between the individual and the State and unwavering trust in the Party State (Zhang et al. 2018).
However, unquestioned loyalty to the State does not seem to translate into the same level of trust within local Party communities, such as fellow student probationers and Party bureaucrats. Huangxu, a 20-year-old urban male student from media studies in a non-key university, discussed his application for Party membership, his experiences of being a probationer and his relationships with fellow probationers and Party bureaucrats: I applied for Party membership because it would be very useful in the job market. But the Party branch bureaucrats are not trustworthy. Their main job is to organise meetings on ideology education. They are game players. They treated us probationers like minions and made us do a lot of chores. They played us against each other, appearing to favour one while supporting another for Party membership. We do not trust each other.
Huangxu's narrative shows a deficit of trust in the Party branch bureaucrats and other student probationers. This confirms previous research on the Confucian model of trust in the State and distrust in the local level of political institutions (Pye 1999;Wu and Wilkes 2018). It seems Party socialisation provides unique opportunities and unifies university students in discussing social policies and public issues. However, whilst this type of political participation might consolidate obligatory relationships between individuals and the State, it seems to be a topdown pattern. The student members or probationers did not voluntarily have regular contact with their Party fellows apart from organised activities. Moreover, this type of engagement did not seem to foster long-lasting bridging or bonding relationships between students. Instead, there seemed to be a lack of trust between student members and probationers.
Furthermore, some students were critical of the purpose of such organised political engagement. Xue Zhang, a 23-year-old urban female postgraduate student, majored in Computer Science in a key university. She had been a Party member for 3 years and participated in a range of activities the Party branch organised during her undergraduate and postgraduate studies. Xue was more critical of this type of political engagement than the other respondents: I have been involved in recruitment. Most new applicants just wanted Party membership as a stamp of approval. They cared about their own interests and joined the Party to advance these interests.
Xue explained that the 'unspoken but widely shared motive' of joining the CCP among university students was to gain access to social or political contacts and/or advance one's own competitiveness. These utilitarian motives which seem to be contrary to the Party members' promotion of the public good were prevalent among Xue's CCP fellows. She further identified the dual pressure from 'degree inflation' and 'the competitive culture' as the underlying 'drive' for students to 'chase' all kinds of academic and non-academic recognition, such as 'outstanding student awards', 'scholarships' and 'useful internships'. Among those 'extra labels', Party membership warrants particular sponsorship, as it 'opens the doors of graduate schools' and leads to 'job opportunities'. However, the pursuit of individual interests seems to alienate the applicants from developing meaningful interpersonal relationships or contributing to the public good. Xue explained that she did not make much effort to socialise with her Party fellows outside Party events, as they were in fact 'competitors' rather than 'comrades'. Xue's account of the competitive culture and degree inflation seems to suggest that individual academic performance is much more valued than students' civic activities in contemporary universities.

Episodic volunteering, generalised trust and bridging relationships
We find a mixed pattern of organised volunteering activities, which are either organised or elicited by the local Party branches or the Youth Leagues on campus. These activities are entirely voluntary, and the students work with some charitable associations devoted to social causes. Nearly a third of the respondents (N = 21) confirmed that they had participated in such voluntary work, but the pattern of their engagement was erratic and most likely to represent a one-off, boxticking activity. For example, Yongzhi Lin, a 22-year-old undergraduate from a medical school, volunteered for the Free Medical Service once. The Free Medical Service Team is an organisation dedicated to providing free dental treatment for the young and the elderly in rural impoverished areas, who are 'out-groups' in terms of interpersonal relationships for Yongzhi: We treated a lot of left-behind kids and elderly. It was an eye-opening experience, but I do not think I could make a difference. This charity is like scattered streams that can never become a river. It should be the responsibilities of their families, right? I would not trust strangers or some organisation to look after my own child or grandparents.
Yongzhi's narrative illustrates the deeply rooted beliefs in family-based welfare and the lack of confidence in long-term charitable commitments, which is consistent with previous research findings (Garrillo et al. 2017). Yongzhi does not believe his volunteering would make any meaningful change for the 'left-behind kids and elderly'. The reference to 'scattered streams' suggests a lack of belief in collective action and an absence of bottom-up civic values, which might also explain the apparent reluctance to make a long-term commitment to volunteering. Furthermore, despite Yongzhi's desire to 'help more people' with his medical skills and knowledge, he could not afford to spend more time with the Team, because of the demands of his studies and the gruelling training at his medical school. He was one of a few medical students who volunteered during the holidays whilst 'other fellow students were busy with their studies or internships in hospitals'. His narrative illustrates the conflicts between the code of meritocracy and moral and social responsibilities. Previous studies suggest that, among students and their parents, there is a persistent belief in the code of meritocracy and upward social mobility through the pathways of higher education (Liu 2016). The belief in merit-based social rewards and selection and the intensified competition at the university level allow students to pursue expressive individualism, thus drawing them further away from committing to civic responsibilities.
Besides organised volunteering activities, we find that small number of students (N = 8) participated in the activities of charitable organisations, including self-organised environment charities and NGOs on safe sex, sexual consent and preventing sexual violence. Chenai Hu, a 24-year-old business postgraduate at a non-key university, has been working for 2 years for a non-governmental charity organisation that addresses domestic violence. Chenai recalled a successful event last November on preventing on-campus domestic violence that was headlined by a popular feminist activist on social media and was live-streamed. Drawing from her business knowledge and contacts, she pushed the charity work to another level by using social media, attracting advertisements, and writing blog posts. However, her main obstacle was to motivate her university peers to volunteer as well: A lot of people think I am a nutter who wastes my time on these things. It is frustrating to persuade my classmates to volunteer their time. When I tried to promote these events, they looked at me as if I was about to trick them into doing something scandalous. Even my own so-called friends do not trust me. I think [university students] are becoming more and more selfish. They do not care about others.
Yuqian and Chenai's difficulty of involving their classmates or university fellows and their frustration with 'selfish university students' are the result of a number of factors, including distrust in their university peers as well as conflicts between civic activities and individual interests. The relentless competition and demands to perform academically, on the one hand, pushed students to be more individualistic in terms of prioritising their own academic pursuits. On the other hand, however, they led students to be satisfied only with measurable successes, such as academic records, grants and prizes. Yuqian and Chenai's characterization of 'selfish university students' thus appears to result from the meritocratic selection and reward process in higher education.

Digital activism and institutional trust
A small number of respondents (N = 9) participated in expressive political or civic engagement, which included writing on-line petitions against professors, celebrities or entrepreneurs involved in sexual harassment scandals, boycotting big business and corporations, and supporting whistle-blowers who exposed environmental, medical or labour scandals. Most of these activities were conducted via social media, particularly via WeChat groups. Most respondents were content with the 'surrogate virtual community', where they held discussions with their 'comrades' rather than meeting them in person. One respondent involved in exposing a sexual harassment case explained: 'You can never be too careful with the current surveillance. You can't trust anyone, especially the university.' A few female students (N = 5) were involved in on-line petitions against sexual harassment by their own professors. All five female students were from urban professional backgrounds. Whilst the recent MeToo movement also spread to Chinese universities, this type of bottom-up activism was not widespread across our sample of respondents. Xicheng Li, a 20-year-old female student from a key university, discussed her experience of on-line petitions against sexual harassment: A famous male professor in our university is alleged to have sexually exploited several female students. It was horrific. But the university has not done anything to investigate. I joined the on-line petition and signed an open letter with my real name. But most of my classmates and friends were reluctant to join in. As we all know, "the gun shoots the bird perching out." We all fear there might be some kind of repercussions. I might regret my actions one day, but I felt it was the morally right thing to do, even if I never had such an experience.
Xicheng further elaborated on the rationale for her petition. Having studied feminist literature and having a professionally successful mother as a role model, Xicheng believes in female empowerment and the importance of standing up against the repression of women. At the same time, however, Xicheng was also aware of possible future censure for becoming involved in on-line activism. 'The gun shoots the bird perching out' is a Chinese saying that describes punishment for being outspoken. Having little trust in her own university, Xicheng seems to be vigilant about possible reprimands, such as losing out in the competition to be nominated an 'outstanding graduate' or 'a graduate school opportunity' at her current university. However, she can afford this possible 'mishap', because her parents will support her financially for a postgraduate degree abroad. Compared to Xicheng's 'confidence', the majority of female students from working class and agricultural backgrounds were cautious about 'putting themselves forward'. Although they agreed that sexual harassment should be addressed by universities, they did not have faith in collective on-line petitions. The risk of being 'singled out' or being seen as 'challenging authority' would cost them everything, particularly since they do not have financial and social resources like Xicheng.

Conclusion
This is a study of how trust affects the nature and characteristics of civic engagement among university students in contemporary China. It has a number of conclusions, some of which relate to the more general literature on trust and civic engagement and some of which are particularly relevant to the Chinese context. First, prior scholarship does not support consistent relationships between levels and radii of social trust and levels of civic engagement. Our research suggests that this is also the case in China in the university context. Generalised trust among students seems to be quite low. The organisations with which they associate either are top-down through the Communist Party, which is ideologically narrow and socially distant among the members, or promote geographical origin-based bonding rather than bridging relationships.
Second, our research suggests that in the university context, typical forms of volunteering have little effect on civic attitudes and behaviour. This finding is at odds with much of the research on volunteering in Western countries, which suggests that volunteering can have long-term effects on civic attitudes and behaviour (particularly forms of service learning) (Campbell 2006). This may be because these activities are predominantly organised top-down in China, and thus, very few of our respondents became involved with charities from a bottomup motivation. This is partly because of their instrumental attitude, which are similar to some of the research on volunteering in European countries, where much of the volunteering is actually motivated by instrumental desires (e.g. to enhance CVs) (Cnaan et al. 2010).
At the contextual level, the unquestioned trust in the State seems to contribute to a clear pattern of organised political engagement, specifically relating to membership in the Chinese Communist Party. Moreover, strong trust in the State also draws university students to value Party membership more than community-based volunteering. Furthermore, our findings confirm the previous research on the lack of civic traditions in Chinese society (Fukuyama 1995;Pye 1999). Whilst there is no precedence in civic engagement to turn to, our respondents seem to have developed a fragile and inconsistent pattern of civic involvement. The majority of associational activities were driven by individuals' interests or particularised trust, which did not seem to connect with the public good. However, a small number of informants developed a sense of moral obligation and public responsibility through voluntary charity work. In spite of this, competition at the university level and the related academic and employment pressure prevented such students from engaging regularly in activities devoted to the public good. The competitive university culture seems to allow expressive individualism to thrive whilst undermining the collective consciousness of the public good and social issues.
Our findings concur with the literature, which demonstrates a high level of trust in the State but lower levels of social trust (Fukuyama 1995;Pye 1999;Huhe 2014). This pattern may be related to the Confucian cultural influence on forms of trust-that is a high level of political trust in relation to the ultimate authority (here, the central State and presumably the CCP) but a low level of trust in relation to local institutions, which is exemplified by our respondents' low trust in party bureaucrats and the way their own universities handle sexual harassment cases.
Moreover, the findings are consistent with the literature on familialism in China and the relatively narrow ambit of social trust in family and kinship networks. This is well illustrated by the prevalence of civic engagement in town-fellowship associations. The findings are therefore in line with the existing literature but add new depth to our understanding of the relationship between the dominant modalities of trust and the typical form of civic participation in the university context. They also add new insights into the effects of increasing meritocratic competition and stress on performativity on the civic attitudes and behaviour of students.
Moreover, the distrust in university Party bureaucrats and fellow Party probationers and/or members does not seem to prevent students from eagerly seeking Party membership. This suggests that utilitarian individualism seems to override the distrust of intermediate Party communities. The unquestioned loyalty and trust in the Party State suggests a persistent ideology of Confucianism and its related moral obligations. On the one hand, persistent Confucianism-related norms and values continue to shape young people's trust and patterns of civic engagement. On the other hand, utilitarian and expressive individualism is fuelled by the competitive culture in contemporary universities. This hybrid of neoliberal individualism and Confucianism seems to be the perfect match to reconcile the absence of civic traditions and to validate civic consciousness among university students.
Regarding individual-level characteristics, the majority of students involved in volunteering and activism were from urban professional families. Among them, there is a slight majority of female students. Due to the sample size, we are unable to explore further the implications of students' family backgrounds and gender on their levels of trust and civic engagement. Although we selected the samples to cover different types of universities and fields of study, our findings do not illustrate clear relationships between the types of universities or fields, the levels of trust and the intensity of civic engagement. These findings suggest further avenues for future research on civic engagement by young people in China. Researchers could explore the impact of young people's socioeconomic characteristics on trust in relation to the patterns of civic engagement. For instance, do the levels of trust vary across different social groups? How do young people from different social backgrounds participate in civic activities? What about those young people from the same age cohorts that lack any higher education experiences? In undertaking such systematic analysis, scholars can sharpen their understanding of the relationships between trust and civic engagement among young people in China in the twenty-first century.