Abstract
This study tackles an important yet marginalized question in the sociology of work: Are there service work-games of resistance? Sociologists kept trying to find them but had little success. Current studies only find service work-games of consent. Building on Lopez’s idea of “successful social interactions” (SSIs), this study searches for service work-games of resistance. The author finds that worker-customer relations resembling SSIs are already reported in some studies, though it remains unclear whether these SSIs are recurrent enough to qualify as work-games and how much they contribute to resistance. Then, the author coins the term work-play to demarcate between ludic informal work routines that resemble play (i.e., work-plays) and those that resemble games (i.e., work-games). After that, the author explains why work-plays have better resistance potential than work-games and why SSIs likely constitute work-plays instead of work-games. On the basis of the qualitative analysis of waitstaff work in a dance club in Beijing, the author found four recurrent SSIs that constitute work-plays and elaborates how these work-plays operated as relational resistance and triggered conventional resistance. These findings confirm the existence of service work-plays of resistance. This study’s data were collected through 19 months of participant observation at a club and interviews with 57 informants.
Keywords
The concept of “work-game” refers to various kinds of work-related games that workers play, including those on the factory shop floor, the service shop floor, and outside the shop floor. Studies on work-games became a central part of industrial sociology with the work of Roy (1959) and Burawoy (1979). The core theoretical issue of work-games studies is whether and how work-games generate consent and facilitate resistance. The sociology of service work has inherited this theoretical focus.
Sociologists’ call for studying service work-games’ consent and resistance implications began in the 1990s. Published in Work and Occupations as a review of the past 10 years’ research in the sociology of service work, McCammon and Griffin (2000) said that
“games” . . . have complex, even contradictory, consequences that reproduce extant power differences in the workplace. . . . Do “games” . . . in the service sector also display contradictory impulses? Too little work has been done to answer such a question with much confidence, but investigation into these matters would be an important way to bridge the literatures on production and service workers and discover what, if anything, is genuinely particular about managerial control strategies and worker accommodation or resistance in each broad employment sector. (pp. 283–84)
Similarly written as a 10-year review of the sociology of service work, Lopez (2010) complains that
the concept of shop-floor games that assumed such a central role in analyses of conflict and consent in studies of manufacturing has been almost entirely neglected in the sociology of service work. . . . The question [is] whether and how the kinds of games service workers play with their customers are different from the kinds of games production workers play with their machines. (pp. 263–64)
Lopez (2010) and McCammon and Griffin (2000) anticipated future studies to find service work-games of resistance and clarify how they operate. However, in contrast to their anticipation, almost all dedicated studies of service work-games find little resistance (e.g., Mears 2015; Sallaz 2009; Sherman 2007). These studies show that service work-games generate consent, serve capitalist interests, and benefit workers in only minor ways. With such findings, the problematic of consent and resistance in service work-games has apparently been settled. Namely, service work-games mainly generate consent, and they have limited potential for resistance.
This study reexamines the problematic and supplies some novel answers. To the question of whether there are service work-games of resistance, this study answers “yes and no.” On one hand, it theoretically shows why work-games, which are a gamelike type of ludic informal work routines, tend to generate consent. On the other hand, it conceptually differentiates between “work-plays” and work-games as two different types of ludic informal work routines. To the question of how service work-games facilitate resistance in practice, this study argues that even though work-games do not tend to facilitate resistance, work-plays do so by constituting “relational resistance” and triggering conventional forms of resistance. These arguments are substantiated by an analysis of waitstaff work in a dance club in Beijing.
Dedicated Studies of Service Work-Games
Like studies of factory work-games, studies of service work-games engage the problematic of worker consent and resistance (Burawoy 2012; Courpasson, Golsorkhi, and Sallaz 2012; Lopez 2010; McCammon and Griffin 2000). Before Burawoy (1979), studies of factory work-games tended to explore their resistance potential. After Burawoy (1979), many studies turned toward uncovering the stealth operation of consent, while a minority continued to identify resistance in factory work-games (Boraman 2017; Durand and Stewart 1998; Koski and Järvensivu 2010; Lin, Lin, and Tseng 2016; Tuckman and Whittall 2002). For example, Koski and Järvensivu (2010) find that work-games helps factory workers block the implementation of neoliberal managerial regimes. This observation is likely generalizable to various industries. Sallaz (2015:6) calls these “resistance games” and cites Vallas (2006) and others as examples.
Almost all current dedicated studies of service work-games stress their consent-generating aspect and doubt their potential for resistance (Lopez 2007; Mears 2015; Payne 2013; Sallaz 2009; Sherman 2007; Vaughn 2009). They find that service work-games feature worker-customer relations that are superficial, inauthentic, economically driven, and/or instrumental. Workers treat customers as a mere “source of unpredictability” or an input for production (Lopez 2010:264). The workers who participate in work-games either aim at gaining incremental economic benefits or try to cope with boredom by using work-games’ gamelike nature. For example, Sallaz (2009:169) identifies an “entrepreneurial game” through which casino dealers hustle to maximize tips from customers and gain the feeling of being entrepreneurs. Sherman (2007:112) finds that hotel service workers play “money games” to hustle for tips and “games of skill and control” to strategically manage customers and gain the feeling of autonomy. Mears (2015) finds that aesthetic workers in high-end dance clubs “play the game of upselling alcohol to clients” (p. 1112). The content of these work-games includes operating under some fixed rules, micro-competitions, wit, smartness, and the psychological pleasures of overpowering competitors and beating the game. In Sallaz’s (2015) scheme, these service work-games are critically characterized as “coping games,” “learning games,” and “reward games” (p. 6).
The only study that stresses service work-games’ resistance potential is Lopez (2010), who theorizes service work-games of resistance as those that feature “successful social interaction” (SSI) between workers and customers. He coined “SSI” as a shorthand term and only briefly discussed it. To my understanding, the concept of SSI should denote worker-customer relations that are to a significant extent noncapitalist, meaningful, affective, and/or voluntary. I fully agree with Lopez that SSIs are a clue to the search for service work-games of resistance. This study uses SSIs as a handle to approach work-plays. However, from a general theoretical perspective, one should not presume that SSIs are the only or necessary component of resistance in all work-games and work-plays. This should be left as an open question for future empirical research.
The Analysis of Service Work-Games from Outside the Research Subfield of Work-Games Studies
Current dedicated studies of service work-games do not find SSIs in work-games. However, I notice that some studies have already found evidence of SSIs. These studies do not cite the term work-game or engage the problematic of consent and resistance. This section discusses these studies. For example, Misra and Walters (2016) observe that “meaningful customer interactions become the main vehicle for [fashion retail] workers to maintain a semblance of their humanity . . . [and] allow workers to find fulfillment and resist managerial expectations to take advantage of customers” (p. 318). Sarpong and Maclean (2017) document how migrant service workers practice ‘service nepotism’ by covertly benefiting coethnic customers at the management’s expense. Besen-Cassino (2014:70–87) finds that youths imagine their service work as an opportunity to practice enjoyable social interactions with customers. A long tradition of research on restaurant waitstaff finds that waitpersons often engage in SSIs with customers and resist managers (Erickson 2004; Gatta 2009; Paules 1991). For example, Gatta (2009) claims that there “are many moments where servers use their agency to maintain a sense of humanity in the work interaction” (p. 80). Hastings and Cumbers (2019:1465) observe a similar kind of interaction in call centers. If one casts a wider net, business management studies on “the bright side of emotional labor” also provide evidence of SSIs (Humphrey, Ashforth, and Diefendorff 2015).
These studies’ discovery of SSIs is invaluable; their findings contradict those of service work-game studies. Nonetheless, I do not think they have proven that there are service work-games of resistance. First, it is because these SSIs do not look like work-games. Burawoy (1985) influentially defines factory work-games as “a set of rules, a set of possible outcomes, and a set of outcome preferences” (p. 38). Workers informing fashion customers about bad deals, practicing service nepotism for coethnic customers, or noninstrumentally interacting with restaurant customers are not gamelike. They are not rule bound or instrumental outcome oriented; they feature no strategic competition or games to be beaten.
Second, whether these SSIs truly facilitate resistance remains uncertain. Some of these studies explicitly argue against interpreting the SSIs in their cases as resistance (e.g., Hastings and Cumbers 2019:1467). Yet other studies that found SSI convincingly present their cases as resistance. For example, restaurant waitpersons are found to help fight against capitalism through various informal interactions with customers (Gatta 2009; Paules 1991). Similarly, retail workers are found to cooperate with customers to beat the capitalist marketing schemes of their stores (Misra and Walters 2016).
Third, it must be stressed that SSIs by themselves do not constitute service work-games unless they are sufficiently recurrent. An implicit part of the definition of work-games is that they must be routine practices. Work-games are informal work routines. For example, the banana time work-game identified by Roy (1959) would not have been a work-game if it had only randomly occurred a few times within a year. The previously cited studies have not ascertained whether the SSIs they found are recurrent or sporadic. Without recurrence, SSIs merely reflect the occasional and random outcome of individual workers’ goodwill. These SSIs constitute a flimsy and unsustainable basis of resistance. However, SSIs become regular work routines with recurrence and, thereby, affect the labor process.
Work-Play in Capitalist Work Contexts: The Analytic Distinction between Game and Play
This section takes a short and necessary detour to gamification studies to theoretically clarify the concept of work-play. Gamification research and its business implementation accelerated in the early 2010s (Faiella and Ricciardi 2015; Mollick and Rothbard 2014). Gamification refers to “the application of game systems—competition, rewards, quantifying player/user behavior—into non-game domains, such as work, productivity and fitness” (Woodcock and Johnson 2018:542). From a critical sociological perspective, gamification may be interpreted as the top-down implementation of work-games to facilitate consent. Sociologists have identified, independently of gamification research, some top-down implemented work-games. Employment games, for example, are top-down work-games for the unemployed (Lopez and Phillips 2019; Sharone 2007). Sharone (2004) also finds management-imposed work-games in creative workplaces.
Lund (2014, 2015, 2017) makes the distinction between game and play to critique gamification. On the basis of ludic theory, Lund (2015:65) interprets games as (1) rule based, (2) goal oriented, (3) competition centered, (4) formally voluntary but often made compulsory by social sanction, and (5) fun in a thrilling way. Lund (2015:65–66) defines play as (1) open ended and continually improvised instead of rule based, (2) targeting intrinsic values rather than extrinsic goals, (3) social communication centered instead of competition centered, (4) proactively initiated, and (5) fun in a leisurely way. Probusiness scholars coined the term playbor to eulogize how digital gamification can turn capitalist labor into enjoyable experiences soon. However, Lund’s differentiation between game and play clarifies that playbor cannot logically exist. When a worker practices capitalist labor, she can hardly be engaging in play at the same time. Gamification can only implement gamelike elements into work. The outcome of gamification should be labeled gamebor rather than playbor.
For this study’s purpose of understanding work-games of resistance, Lund’s interpretation helps by illustrating the logical possibility of mixing play with work. Deductively speaking, this mixing should result in noncapitalist or anticapitalist characteristics because play is inherently noninstrumental. Work has been heavily corrupted by capitalism in contemporary society; capitalist work is characterized by exploitation and alienation. However, because work is not inherently or necessarily capitalist, it is not inconceivable in theory that play and work can be fused.
The previous discussion enables us to differentiate between work-games that are gamelike, which are conventionally labeled work-games, and work-games that are playlike, which should be given a new label that highlights their playlike characteristics. The new label that I proposed is work-play. Lund (2014) calls the outcome of mixing play and work playwork, though he does not further develop the concept. I renamed it work-play to connect it to work-game studies.
I clarify that the distinction between work-games and work-plays is analytic. Work-plays and work-games are inevitably ideal-typical concepts, simply because game and play are ideal-typical concepts. Actual instances of games and plays are not perfectly ideal-typical. They occupy various positions along a continuous spectrum on which ideal-typical work-plays and ideal-typical work-games designate the two ends. In reality, some work-games can demonstrate certain playlike characteristics and vice versa.
Ludic Informal Work Routines: The General Category to Which Work-Game and Work-Play Belong
Given the differentiation between work-games and work-plays, it makes sense to coin a new term for the general category that covers work-games, work-plays, and perhaps other relevant subcategories that future studies will identify. Should playlike work-games, or work-plays, not be included as a part of the research object of work-games studies? If the answer is positive, then this new term can also be used to update the conventional naming of work-games studies.
I coined the term ludic informal work routines to represent the general category. I define ludic informal work routines as parts of work processes that are (1) informal (as opposed to being institutionalized by the management), (2) relatively recurrent, and (3) at least modestly enjoyable in gamelike or playlike ways. One can view work-games as ludic informal work routines that centrally feature gamelike elements, and work-plays as ludic informal work routines that centrally feature playlike elements. Note that the term ludic informal work routines does not preempirically specify SSI’s functioning or delve into the problematic of consent and resistance.
The terms work-play and ludic informal work routines benefit work-games studies in three ways. First, and obviously, they delineate an expanded research field for work-games studies. The research object becomes broader as it encompasses work-games, work-plays, and possibly other unexplored types of ludic informal work routines. They advance the research subfield by cultivating theoretical space for work-plays vis-à-vis the established concept of work-games.
Second, the two terms help clarify the critical assessment of work-games initiated by Burawoy (1979). It helps explain why both factory work-games and service work-games often facilitate consent. Because of games’ instrumental, competitive, rule-based, and compulsory characteristics, they are not geared toward encouraging noninstrumental worker-customer relations. Because of their gamelike characteristics, work-games tend to be incongruent with playlike ludic practices at work.
Third, the two terms help rectify a theoretical misstep in the sociology of service work. The current conception of service work-games inherits from that of factory work-games without significant revisions or independent theorization (e.g., Halpin 2013; Sherman 2007:111). However, such inheritance should have been regarded as suspicious from the perspective of the sociology of service work, which is partly founded on the belief that service work considerably differs from manufacturing work.
The conventional conception of factory work-games presumes that all ludic informal work routines in capitalist work must be gamelike. Alternatively put, it presumes that games are the only socially relevant ludic practice in capitalist work. The proven existence of SSIs and this study’s analysis of work-play necessitate the reconsideration of this presumption. Work-plays illustrate that ludic informal work routines in factories and service workplaces can considerably differ. Work-plays’ existence supports sociologists who maintain that service work differs from factory work because of worker-customer relations (Bélanger and Edwards 2013; Havard, Rorive, and Sobczak 2009). The existence of service work-plays of resistance corroborates the argument that the analysis of worker-customer relations enhances rather than obstructs the critical analysis of service work (Brook 2007; Korczynski 2007, 2013; Warhurst, Thompson, and Nickson 2008).
Relational Resistance
Conventional worker resistance assumes various forms that range from industrial action to worker-led armed rebellions. I call it “conventional” because it is the most scholarly recognized type of worker resistance. It is nondiscursive, nonsymbolic, overt, collective, structural, and explicitly oppositional. As research on resistance evolved, an additional and different type of resistance gradually became recognized. Such worker resistance is discursive, cultural-symbolic, covert, and often individualized. Many scholars initially denied that it constituted resistance. However, recent scholars define resistance sufficiently inclusively and explicitly recognize cultural resistance as resistance (Baaz et al. 2016; Courpasson and Vallas 2016; Mumby et al. 2017; Thompson 2016).
At first glance, SSIs may seem like cultural resistance. They obviously differ from conventional resistance. Like cultural resistance, all SSIs investigated in current studies tend to be individualized instead of collectively coordinated and carried out. Play and interactions are more readily seen as cultural and hidden instead of structural and explicitly oppositional. But the resemblance terminates here.
As shown in current studies, SSIs do not primarily concern symbols or discourses; they are not conducted through symbolic or discursive means. Moreover, with some exceptions, they are not hidden and secretive in the ways that cultural resistance is. Some of them are even openly performed and/or well known by managers, as this study’s analysis will show. Furthermore, SSIs are composed of workers’ social interactional practices with customers, whereas cultural resistance is usually composed of workers’ individualistic textual and discursive practices.
So which type of resistance do SSIs potentially constitute? On the basis of observations during fieldwork and associated grounded theory processes, I arrived at “relational resistance” as an answer. Later sections will empirically substantiate this answer. The rest of this section reviews the theoretical background of relational resistance. Relational resistance already looks like a plausible answer from a deductive perspective. As micro-level social interactions, SSIs are inherently relational. Workers’ constructing noncapitalist social relations with customers is in itself relational-processual. SSIs are by definition noncapitalist worker-customer relations; recurrent SSIs necessarily displace capitalist worker-customer relations to some extent. Even though SSIs only briefly appear in the service workplace, they practically displace capitalist worker-customer relations in these brief moments. This displacement is not merely symbolic or performative.
Relational resistance is not (yet) a highly recognized terminology in sociology and resistance studies. Originally coined by psychologists and others in the late twentieth century, the concept has only become increasingly used in the past several years (Celani 1998). It is cited by 260 documents between 2015 and mid-2025 according to Google Scholar. It is cited mostly in psychology, education studies, geography, business studies, organization studies, and feminist studies (Bear and Holloway 2019; Fawcett et al. 2015; Marsh 2024; Saleh 2019; Wilhoit and Kisselburgh 2019). Only a few sociologists have cited it (Belisle and Dixon 2020; Maye-Banbury 2015; Rosenbloom 2004). Most of these studies use the term in casual, descriptive, or nontechnical ways, as if its meaning is self-evident (e.g., Maye-Banbury 2015). The rare attempts to theorize it tend to be weak and brief (Bear and Holloway 2019; Marsh 2024; Serafini 2018). Sociologists and resistance studies have much further work to do to theorize this concept.
Despite using the term in moderately different ways, current studies converge in conceptualizing relational resistance as sociopolitical resistance that is primarily relational. Examples include teachers and students forming deep relations to resist racial discrimination, or agents resisting the cooperation between different business firms. This study adopts this general conception of relational resistance to interpret SSIs and work-plays. This conception does not claim that relational resistance stands with cultural and conventional resistance as three major types of resistance; it does not argue that all resistance is relational (Wilhoit and Kisselburgh 2019).
Sociological Research on Nightlife
This study primarily tackles the puzzle of service work-games of resistance. It does not specifically address the emerging interdisciplinary research field on nightlife (Acuto 2019; Nofre and Ruiz 2023). Meanwhile, this study is a dedicated sociological analysis of (work in) nightlife. In this capacity, it indeed contributes useful new knowledge to the interdisciplinary field of nightlife. This section outlines this contribution. I describe sociologists’ main approach to nightlife studies as “nocturnal sociology,” which in turn contributes to the emerging vision of “nocturnal social science” (Shaw 2015). Nocturnal sociology began with Melbin’s (1978) seminal work, which conceptualizes nightlife as an unexplored social frontier that may be analyzed through a wide variety of sociological perspectives. However, sociologists have been much less interested in nocturnal social science than geographers, cultural studies scholars, and others. As Mears (2016) astutely observed, sociological studies of nightlife are few, and they have not extended the investigative scope beyond racial and gender inequalities in nightlife (Grazian 2008; May 2014).
Work and workers in nightlife are clearly a research area that nocturnal sociology should explore. The call for attention to night-work has been made by geographers, sociologists, and cultural studies scholars (Chatterton and Hollands 2003; McRobbie 2002; Shaw 2022). Few sociologists answered the call. The ones that do so examine relatively privileged night-workers such as bouncers, bartenders, managers, affective workers, and hostesses (Coffey et al. 2023; Hoang 2015; Hobbs et al. 2003; Mears 2020; Threadgold, Farrugia, and Coffey 2021). This study contributes to nocturnal sociology as the first dedicated article on frontline manual service workers in nightlife. Additionally, approaching night-work from work-games and worker resistance perspectives, it helps demarcate the nocturnal sociology of night-work from mainstream approaches. Economics-oriented and urban policy approaches influentially imagine night-work in terms of the geography of nighttime economy, and thereby occlude frontline manual service workers from scholarly attention.
Methods and Data
This study’s data were collected through participant observation, in-depth formal interviews, and informal ethnographic interviews. Data collection was conducted by me and a full-time research assistant, Chris. I collected data by myself and directly from the field between June and August 2014, between June and August 2015, August 2016, and June 2017. Chris conducted additional interviews and worked as a recording assistant for participant observation between September 2014 and August 2015. This data collection was for my larger research project on dance club workers in China. I already had much experience of observing the customers, business operations, consumer culture, and service work in dance clubs in different Chinese cities when I started this research project in 2014 (Chew 2009a).
My primary field site was an upper-market dance club, DG. I visited the main dance club district of Beijing at least six nights a week. I stayed in the club for between one and five hours each night. I informally chatted with waitstaff and observed their work. Through working as a waitress at DG, Chris arranged for me to interact with the waitstaff, other frontline staff, and managers. Both Chris and I went with participants to postwork meals, went shopping with them, and hung out with them on their holidays. She participated in all routine work and leisure activities of coworkers, including working in the dance club workplace, living in the company dormitory, having postwork meals, feasting on paydays, and chatting online with waitpersons from DG and other Beijing clubs.
Chris and I conducted informal ethnographic interviews with all waitpersons working in DG. Table 1 shows their home provinces, age, gender, and length of work experience at DG. We arranged multiple sessions of formal interviews for 45 participants. We conducted multiple sessions of informal interviews for 12 participants who disliked the atmosphere of formal interviews. The informal interview settings were conversations over postwork meals, hangouts during holidays, and online communication. Formal interview settings included cafes, restaurants, and the company dormitory. I also informally interviewed some middle-level and high-level managers; 4 were from DG and 11 from other clubs.
A List of the Main Informants and Their Relevant Information.
In January and February 2026, I conducted a small-scale follow-up round of data collection. The goal was to update my findings and check whether the four work-plays that I found in my main dataset (which was collected between 2014 and 2017) are still present in 2026. I report this update in the “Conclusion” section. I carried out semistructured interviews with six individuals, including a middle-level manager who used to work at DG, the general manager of a dance club in Shenzhen, two waiters located in Beijing (who had not worked at DG), and two veteran clubbers located in Beijing. I also broadly read various online sources that contain information on nightlife transformations in Beijing and China between 2019 and the present.
The design and practice of my data collection were informed by phenomenological methodology. Although conventional single-round formal interviewing is a valid method, it severely limits the depth and thickness of interview data (Lamont and Swidler 2014). The design of my interview partly resembled phenomenological hermeneutics, which “aspires to understand other people’s life-worlds more profoundly than the widely accepted research practice of treating interview transcripts as data” and “uses narrative interviews only as a tool to organize subjectively recollected experiences” (Eberle 2015:563, 577). My participant observation practices borrowed from lifeworld analytical ethnography (Honer and Hitzler 2015). Lifeworld analytical ethnographers
prefer an ethnographic approach where the researcher stays in a field for a considerable amount of time, observes sequences of actions repeatedly and talks to the same people many times and in the context of their actions (e.g., their work setting and not someplace else). (Eberle 2022:120–21)
The major rounds of formal interviews began in March 2015. I prepared for these interviews after six months of pilot informal interviews and participant observation between July 2014 and January 2015. The interviews were semi-open-ended. On one hand, the entire first-round interview relied on several prompts instead of a large number of standardized questions (Jiménez and Orozco 2021). This gave the interviewees more room to initiate the discussion of details, posing counter-questions, and leading the interview toward relevant tangential directions. On the other hand, I set up prompts that centered on the four SSIs that I discovered between July 2014 and February 2015. For example, one of the prompts was “Can you describe to me your experiences of partying with customers?” In second-round and third-round interviews, I asked follow-up questions and obtained a more in-depth understanding of the lived experience of each interviewee.
My larger research project on dance club workers in China commenced in 2014 with a grounded theory methods orientation (Timonen, Foley, and Conlon 2018). During the first three months of data collection, I completely refrained from adopting any theoretical frameworks to analyze the data or build theory. After the third month, I gradually arrived at five themes to which my larger project potentially contributes: work-games, boundary work, subcultural style, worker resistance, and consumer-worker identity. I treated these five themes as the basis for five separate journal article projects. (A few of them were already published.) For this article, an abductive element was introduced into the grounded theory process (Tavory and Timmermans 2014). As explained in previous sections on theory, I read the work-games scholarship and noted its unsuccessful search for service work-games of resistance. Hence, SSIs were an anomaly. One of the directions in my early-stage participant observation was to look for SSIs in the field site. I experimented with various frames to interpret what I observed. I made comparisons between different clubs, waitpersons, and informal work practices. These processes occurred between late 2014 and late 2016. I also conducted theoretical sampling and reinterviewed informants with new prompts. After reflecting on the data many times between 2016 and 2018, I eventually settled on an interpretation of the SSIs in terms of relational resistance.
Finally, I outline positionality issues in the data collection process. Guided by phenomenological methods, I started this process with a very conscious effort to ameliorate the potential effects of researcher positionality. My education, class background, and age greatly differed from those of my informants, creating substantial social distance. They could be compelled to give me answers that are more polite and socially acceptable than what they truly thought. This is a very common type of positionality issue that involves qualitative interviews with socially underprivileged groups. I dealt with it by relating to all informants humbly and ethically. For example, because I supplied informants with intellectual and cultural knowledge on nightlife, they felt glad and rewarded about participating in this research. Another minor positionality issue concerns my ethnicity as a Hong Kong citizen. All except three of my informants were officially defined as Han Chinese on their government-issued identity cards. The other three looked much like Han Chinese despite their official non-Han classification. My informants treated me as a partial outsider of the Han Chinese ethnicity. I felt that this issue did not practically undermine data collection; it actually encouraged informants to give me more context.
Besides the phenomenological deepening of the interviewing, the single biggest research design feature I implemented to ameliorate positionality effects was the strategic choice of my research assistant. Chris is cisgender female. She was 21 years old in 2014 and had just graduated from a university. According to Chris, her class background was “between middle class and lower middle class.” She came from a nonwealthy province and a prefecture-level city. In terms of class, place of origin, education, and age, she was moderately close to the informants. She and I compared notes and discussed ways to approach specific informants for in-depth information. Meanwhile, her gender could have created two obstacles: (1) “getting into the shoes” of male informants and (2) collecting information on sex related issues from male informants. She handled the first obstacle and largely resolved it by, for example, gaining respect from male workers for working very hard. The second obstacle was not easily resolvable. But because my project does not focus on sex in the first place, I settled for less detailed data on sexuality issues.
Waitstaff Service Work in a Dance Club in Beijing
China’s nighttime economy is enormous, complex, and locally distinctive. General information on it is available in Chew (2009b, 2026) and Farrer (2018). Dance clubs and bars, which are Western nightlife institutions, occupy the highest profile position in the ecology of Chinese nightlife institutions. But local institutions, including karaoke bars, hostess nightclubs, and foot massage parlors, are no less large scale and important. Although dance clubs in China superficially look like those in Western cities, they are considerably different because of cultural localization (Chew 2020). A localized feature that matters to this study is that service workers often intermediate and intervene in customers’ consumption experiences. Like dance clubs’ waitstaff in the United States, those in China bus tables and do the cleaning up. However, they also regularly chat, play, and interact with customers. These localized service work practices are relatively less pronounced in China’s southern coastal cities. Elsewhere, including Beijing, they are firmly established.
In Chinese dance clubs’ pay structure, waitpersons earn a large portion of their income from tips. The base monthly wages at DG were 1,200 yuan, but the average monthly income of my informants was about 5,500 yuan. (The exchange rate was 1 U.S. dollar to about 6 yuan in the mid-2010s). They did not earn commission for selling alcohol. Only sales staff (xiaoshou), a different kind of frontline service workers, did. With this pay structure, the waitpersons tended to develop a feeling of autonomy (Brewster and Jeremiah 2013). They also tended to perceive their relations with customers as equally important as their relations with managers. The managers were not focused on maximizing output from the waitstaff with carefully designed and regimented labor processes. Hence, dance club labor processes were considerably less structured than those in typical Chinese factories and service workplaces such as restaurants. Overall, Chinese dance clubs featured labor processes and pay structures that were moderately conducive to SSIs.
Meanwhile, the waitstaff labor at DG was no less exploitative than in other workplaces in China. Waitpersons’ formal responsibilities included bussing the several tables assigned to them, continuously serving customers at their tables, cleaning up after customers leave, and creating buzz for the club through acts such as dancing. Emotional and aesthetic labor were required. Waitpersons worked in cigarette smoke-filled spaces from around 6 p.m. to 5 a.m. They only had one meal break that lasted 20 minutes. On the three or four nonbusy days of the week, workers could leave at about 2:30 a.m. They had three or four days off per month. A surveillance team of five, which operated independently of security guards, was present to monitor the actions of all waitpersons. Although the primary objective of the monitoring was to regulate the waitstaff’s handling of tips and bills, other waitstaff behaviors were also monitored. I did not observe or hear about the sexual exploitation of waitresses and waiters by customers. Other specialized service workers were present in some dance clubs to cater to customers’ commercial sex demands.
The First SSI: Copartying with Customers and Partying with Coworkers
Identifying and Analyzing the First SSI
Fashion retail stores, theme parks, high-end restaurants, and cafés offer their frontline service workers an enjoyable and/or stylish buzz (Silver and Clark 2016). Dance clubs offer an even more immersive and aesthetically rich buzz. There are live DJ sessions, fashionable people, mesmerizing lighting, and dancing. Specialized creative workers are employed by dance clubs to help produce these different aspects of the buzz. Because of the previously mentioned localized work practices, club waitpersons in China routinely contribute to producing this buzz. In this context, these workers readily introduce play into work. This section examines two aspects of such play: service workers’ copartying with customers and their partying among themselves. Service workers in clubs and bars in Western societies, unless they are hired primarily as affective workers, hardly engage in such play (Threadgold et al. 2021). They mainly bus tables; Western customers are not used to partying with waitpersons. In contrast, Chinese clubs hire a larger number of waitpersons to share the manual labor, and Chinese customers are used to partying with waitpersons.
Chris and I observed that as our informants copartied, they transformed themselves and the micro-situation in noninstrumental directions. We asked informants to reflect on how they managed or decided to start copartying. Informant 28 answered that his “shift from a work mode into a play mode was natural.” Informant 20 said that as he started to party or coparty, he “tended to forget about oneself and just focus on the dance and music.” Informant 9 described how she “forgot about her identity as a worker” during these moments of partying. Informant 38 said that she “can’t help swaying her body and dancing once she hears a favorite track.” I interpreted these answers as a vivid phenomenological illustration of the emergent, noninstrumental, and unplanned nature of work-play. They also demonstrated that the relational transformation—from capitalist worker-customer relations to noninstrumental and nonhierarchical relations among individuals in play—was both effortless and effective.
My informants’ copartying with returning customers especially deviated from the typical capitalist worker-customer relations. The transformed relations could become quite strong because of prior copartying experiences and already established acquaintances. The copartying between waitpersons and customers could occur at various times throughout a club night. My informants danced to the beat as they bused tables, danced with customers for a short while, and drank a round with the table of customers they served. Copartying can hardly be interpreted as an instrumental work-game or a work-game of consent. A tipping work-game was established among the waitstaff. Informant 45 informed me that its central principle was “to prioritize creating the impression of performing onerous labor.” Partying with customers or coworkers was considered by my informants to be counterproductive for creating this impression: copartying discouraged customers from seeing a waitperson as someone performing onerous labor.
My informants recurrently partied among themselves. During nonbusy days and nonbusy hours of busy days, they did not need to perform any manual labor. I observed that they often grouped in a circle and danced, joked, played drinking games such as “dice” (without drinking), and enjoyed the music. During busy days or hours, they still developed ways to party among themselves. I observed that they used their torchlights to send light signals to each other in rhythmic ways as they stood by the table they served. Informant 25 said the waitstaff gleefully labeled this practice “making use of light to communicate love with each other.” They casually danced along the way as they transported drinks from the bar to customers. They did high fives and other secret handshakes as their paths crossed. Note that partying with coworkers is not a social interaction based on worker-customer relations. Therefore, strictly speaking, it is not an SSI. However, it still qualifies as work-play if it occurs recurrently; it is a work-play based on worker-worker relations.
Are waitpersons’ partying and copartying emotional labor (which is a part of the formal work requirements) or an instrumental work-game that maximizes tips? My data did not support these two interpretations. Chris and I observed on many occasions that the informants genuinely immersed themselves in the music and the buzz. If their partying and copartying indeed produced similar effects of emotional labor, then these emotions were not alienated, and they were not a part of the formal work routine.
Night-work in dance clubs and bars is highly gendered (Chew 2019; Coffey et al. 2023; Threadgold et al. 2021). Female service workers in the Chinese nighttime economy also suffer from mainstream society’s stigmatization of their occupation as resembling prostitution (Chew 2019). Despite gendered differences in the work environment, most of my female informants reported to me their participation in the first SSI. They partied with coworkers as frequently as male workers do, though they copartied with customers less frequently than male workers do. As informant 48 explained, she “felt comfortable copartying with customers only when a table largely consisted of female customers or was headed by a female customer.” Because such female-led groups of clubbers are fewer than male-led ones, she did not coparty much. Meanwhile, my female informants showed no social inhibitions in joining male workers in partying. Informant 8 thought that she “cherished the rare occasions in which she partied with a group of female-only workers,” but “felt as enjoyable partying with a mixed gender group of workers.”
The First SSI’s Resistance Implications
The first SSI’s resistance implication may be understood in terms of two relational transformations in the workplace. First, the SSI introduces a transformation in worker-customer relations. Although this transformation is momentary, it solidly goes in an anticapitalist direction. When workers and customers coparty, they to some extent replace capitalist worker-customer relations with noninstrumental relations between the playmates. The new relations are more convivial, nonhierarchical, and aesthetic oriented than the original capitalist worker-customer relations.
This relational transformation could appear tame and/or unimpactful to an observer who is looking for direct confrontations. However, with deeper reflection, one can find two ways in which such relational transformation involves confrontations. The first way concerns workers’ conflict with customers who do not embrace the relational transformation. Some customers disliked it because they were used to the impersonal worker-customer relations found in clubs in Western societies. For these customers, the new relations could not materialize. When my informants served a table consisting mostly of such customers, they withheld their copartying attempts. My informants and other waitstaff at DG identified a quite large group of customers who disliked the transformation, “foreigner-guests” who were invited to party at DG free of charge. In a subsequent section, I will document how the waitstaff protested against foreigner-guests, terminated this group’s privileges, and practically expelled them from DG for good.
The second way concerns workers’ conflict with the management. On the basis of my interviews with DG’s managers, they truly expected the waitstaff to behave according to the formal routines. In general, formal work routines are strictly implemented in contemporary Chinese factories’ labor process; contemporary Chinese workers endorse a work culture of obedience (Chan and Selden 2014; Ngai and Smith 2007). A middle-level manager of DGs complained to me that he “wanted waitpersons to dance only when there were few customers” or “when the buzz was lacking.” My informants partied among themselves and copartied with customers much more than the managers desired. These managers failed to regulate the waitstaff only because “the waitstaff’s demand for having fun dancing was strong.” I observed similar latent conflicts in other clubs across China.
The second resistance implication of the first SSI is its transformation of the relations between workers and (tangible and intangible) inanimate things in the workplace. This transformation goes toward a noncapitalist direction. The inanimate things include the sound system, the lighting, the interior decoration, the buzz, and others. I showed that my informants came to relate to these things as consumers. Originally, these inanimate things were means of production at the service workplace; they resemble machines in the capitalist factory. The first SSI turned these machines into workers’ means of play and consumption. As Burawoy (1979) notes, factory workers develop work-games that involve machines. My analysis shows that ludic informal service work routines may also involve inanimate means of production.
This section has established that the first SSI that I found at DG qualifies as a work-play of resistance. It is very recurrent. My informants copartied with customers and partied among themselves almost nightly. I also revealed two ways in which this SSI transformed worker-customer relations toward noncapitalist orientations.
The Second SSI: Sharing Cultural and Material Resources of Clubbing with Customers
Identifying and Analyzing the Second SSI
The second SSI that I found at DG was based on my informants’ relating to customers as selfless sharers of cultural and material resources of clubbing. At least a significant minority of customers of DG and other high-end dance clubs in China were not knowledgeable clubbers. Instead, they were middle-aged people who engaged in clubbing as a part of their middle-class or wealthy lifestyle. They neither understood electronic dance music nor loved dancing. On one hand, my informants—youths with a significant amount of clubcultural capital and experience—adopted the viewpoint of regular clubbers and secretly despised these “occasional clubbers” (Chew 2019). On the other hand, they also acted as club professionals who earnestly helped transform these occasional clubbers into knowledgeable ones.
I observed that my informants verbally explained to occasional clubbers about the club scene, regular activities of a night out, and fads in drinks and dance. Informant 29 complained that “this task was tedious and often interrupted by manual labor and loud music.” Informant 12, who was at the interview site with informant 29, responded that he felt an urge to educate the occasional clubbers: “It was like seeing little kids giving totally wrong answers to a simple mathematical question; one would feel the responsibility to teach the kid to do it the right way.” Their attitude reflected a noncapitalist and noninstrumental orientation.
I observed that my informants instructed customers how to dance to the beat, act cool in a club, and behave like knowledgeable clubbers with clubcultural capital. They did it through verbal and nonverbal means such as body language. For example, most occasional clubbers did not know American-style secret handshakes but thought they were cool. A few of my informants had the experience of teaching clubbers to do such handshakes. Reflecting a noncapitalist orientation, a part of this education was against the economic interest of the waitstaff themselves and the club managers. An example was educating customers on the comparative consumer value of different clubs. DG was not the most cost-effective and culturally authentic among clubs in Beijing’s nightlife district. My informants did not refrain from honestly telling clubbers about this.
Besides cultural knowledge and know-how, my informants gave considerable material resources to customers. Managers of DG gave each waitperson a small allowance of free food and drinks for them to offer to customers. It was a common marketing arrangement of Chinese dance clubs. It did not represent SSIs to the extent that it was formally planned by managers. However, to the extent that my informants practiced service nepotism or went beyond the allowance limit, they were practicing SSI. As informant 31 testified, he often “offered free food and drinks to favorite customers rather than DG’s intended targets.” Informant 40 used to work in the kitchen and had good relations with the cooks. She “was able to obtain extra food” and “often did it for her favorite customers.”
The most relationally impactful material resource that my informants shared was not food and drinks. It was free entrance to high-end clubs and a free table in these clubs. There was a semiformal system that allowed waitstaff of all dance clubs in Beijing’s nightlife district to freely enter other clubs, bring with them a few guests, occupy a table, and receive selected items of free food and drinks there. The total economic value easily added up to several hundred U.S. dollars per night out, an amount that rivaled the average monthly wages of fresh college graduates in China. My informants and club workers in other Beijing clubs eagerly and tactically exploited this system. They offered their favorite customers more freebies than what the system formally authorized them. For example, they got admission to a club more times than specified, brought numerous customers as free guests into the club, and drank more drinks than allowed. This was possible because the actual gatekeepers of these freebies were their coworkers, low-level managers, or friendly waitpersons of another club. These individuals collectively undermined the system’s semiformal rules. My informants made ample use of this resource and shared it with their favorite customers at least once per week.
I observed that my female informants participated in the second SSI as fully as my male informants did. Most of them enthusiastically secured free entrance to clubs and free food and drinks. In some respects, they ran free tables in various clubs in even more efficient ways than my male informants. For example, informant 27 was more sociable and interacted with a wider range of clubbers than my male informants. As she testified, “The rewards of organizing these free tables were substantial. . . . The free food, drinks, and entrance are secondary. The chance for friendship, fun, and networks, and the pride of hosting a memorable night out are invaluable.” A minority of informants, both male and female, did not fully exploit these resources because of their introverted personality, moodiness, and subpar health conditions. For example, they organized such free parties only once a month instead of once per week.
The Second SSI’s Resistance Implications
The second SSI’s resistance may be interpreted in terms of two relational transformations. The first relational transformation resembles that of the first SSI. When workers share their cultural and material resources with customers, they substituted capitalist worker-customer relations with noninstrumental, nonhierarchical, noncapitalist, and affective relations. Capitalist worker-customer relations were solidly displaced. Incidentally, this relational transformation is exactly what is emphasized in current studies that find SSIs (Gatta 2009; Hastings and Cumbers 2019; Misra and Walters 2016; Paules 1991). It is also the basic assumption of Lopez’s (2010) heuristic conception of SSI. My findings on the second SSI directly support one of the foundational arguments of the sociology of service work: an investigative focus on worker-customer relations enhances rather than obstructs the critical analysis of service work.
The second SSI additionally constructs a noncapitalist relational structure in the capitalist service workplace. This is the second relational transformation that the second SSI brings. As gatekeepers of the freebie system, my informants collectively conspired to bypass the system’s semiformal rules and amply shared resources with their favorite customers as friends. They built a noncapitalist relational structure, and it was parasitic on DG’s capitalist operations. It is built through the recurring relational practices of turning customers into friends and defying capitalist employment relations. This relational structure is not only noncapitalist but also anticapitalist. Controlled and run by waitpersons, it advances the social imperatives of workers and customers instead of the capitalist imperative of profit-making. I participated in this relational structure and personally benefited from it. I recorded in my fieldnotes an informative comment made by a customer of informant 36. This was a senior-year university male student; he felt that the free parties organized by his waitstaff friend were more “real” than regular clubbing. He highlighted the genuinely human, unalienated, friendly, and nonhierarchical contexts of these parties. His feelings reflected that the worker-generated relational structure practically facilitated SSIs.
This section has established that the second SSI qualifies as a work-play of resistance. The sharing of cultural resources occurred at least a few times a week. The sharing of material resources occurred at least once every few days; it occurred when the chance came up. The organization of free tables and parties occurs once per week for most informants. This section highlighted a way in which the second SSI transformed worker-customer relations toward a noncapitalist orientation and another way in which this SSI created an anticapitalist relational structure.
The Third SSI: Building Weak Social Ties with Customers
Identifying and Analyzing the Third SSI
The third SSI emerged through my informants’ efforts to build weak social ties with middle-class and wealthy urban customers. The broader social context of such efforts is the rural-urban divide and the treatment of rural migrants in China (Keung Wong et al. 2007; Solinger 1999). Younger generations of rural migrants, or those born after the 1980s, ultimately wish to be socially and culturally integrated into urban China. They do not wish to return to their rural homes when they grow old. Club waitstaff work offers them a relatively rare pathway to such social and cultural integration.
Informant 25 explained, “I chose the occupation of club waitstaff because it would help broaden my horizons in the city. Other occupations do not give me the chance of routinely connecting with urbanites.” Informant 44 said that “those who did not strongly want integration would have chosen a daytime occupation instead; the wages were similar.” Club waitstaff work practically requires workers to live during the nighttime and sleep during the daytime; this causes physical and/or social hardship for some. For example, informant 14 said that he “suffered from health problems and lacked regular social life, but would still stay in DG because this job was the key to longer-term urban integration.”
Similar to these three informants, others emphasized that broadening horizons and urban integration were the major reasons for them to work at DG. An interesting case was informant 10. Although his residency was rural, his family was middle class because of the successful operation of a small local business. Being middle class, informant 10 had the economic resources to ensure a permanent stay in the city. Yet he still felt the need to work as a waitperson at DG. He reasoned that “compared to all other workplaces, clubs afforded more opportunities for socializing with urbanites.” I could understand why he thought so. Waitpersons and customers are physically close to each other in the club, such as when they talk over loud music. Typical nights out are lengthy; they last at least a few hours. Waitpersons in Beijing serve only a few tables per night.
The previous discussion reveals that my informant’s actions of nurturing weak ties with customers are purposeful. Given their purposeful nature, do these actions and the resultant tie constitute an SSI? First of all, the tie and the actions were noncapitalist; it helped displace capitalist worker-customer relations. Moreover, although these actions were purposeful, they were not exactly instrumental. My informants hope to use weak ties with customers as the micro-interactional basis of social and cultural integration. For example, informant 56 “wanted to learn the manners and the know-how of being a regular member of the metropolis.” The weak ties were not merely interpretable in terms of aspirational capital and social capital; they let my informants internalize the implicit culture of metropolitan people (Lizardo 2022).
I observed weak ties formed between regular customers and their waitpersons. Some regular customers would book in advance and designate their preferred waitpersons before they visited a club. Some instances of these weak ties operated to assist the waitstaff beyond facilitating social and cultural integration. A veteran manager informed me, “When Beijing’s waitpersons decided to quit working in the nighttime economy, quite a few of them relied on guanxi with customers to secure daytime jobs.” Informant 25 said that “a customer had asked him to leave DG and join the customer’s restaurant as a daytime service worker.”
I observed that my female workers similar wish to build weak ties with customers. But there were subtle differences in practices because of the heavily gendered worker-customer relations in night-work. I found that three female informants were more hesitant to build such ties than the average male informant. Informant 15 explained that, as a rule, she “only built ties with female customers due to possible sexual and romantic misunderstandings with male customers.” Meanwhile, two female informants managed to build stronger ties with customers than the average male informant did. Informant 27 had an amiable attitude and good work abilities. After building ties with various customers, she got employed by a customer and became a midlevel manager at a bar in Beijing’s nightlife district.
The Third SSI’s Resistance Implications
The third SSI partly differed from the first two SSIs because of its intersectional impact. Like the first two SSIs, it facilitated new worker-customer relations that directly displaced capitalist worker-customer relations. But unlike them, it additionally helps Chinese rural migrant workers resist the rural-urban divide. It did so by cultivating new social relations outside the workplace. It cultivated weak ties that facilitate rural migrants’ social and cultural integration. It occasionally developed into longer term guanxi that operated in the daytime and outside the nightlife context. Guanxi, which refers to Chinese-style relations, differs from Western relations in its emphasis on entwined instrumental, moral, and emotional elements. Even weak guanxi can yield useful benefits, such as referrals to jobs.
The third SSI’s intersectional implication deserves more contextualization. Although resistance against the rural-urban divide and resistance against capitalism are analytically distinct, they are inextricably intertwined in contemporary China. Beginning from the Maoist era, the Chinese state set up various social policies, such as the notorious hukou system, to separate China’s rural and urban population (Kuang and Liu 2012; Zhou and Zhu 2019). Because of this unique political background, China’s rural-urban divide has a dual structure. On one hand, it is a socially oriented and bottom-up-driven type of oppression that emphasizes racialization, cultural discrimination, and microaggression (Chew 2019; Han 2010). On the other hand, it is a politically invented and top-down-driven type of oppression that emphasizes formal stratification, differential citizen rights, and physical urban exclusion (Li, Gu, and Zhang 2015; Wu and Wang 2014). Despite China’s rapid economic development and urbanization, such dual oppression persists even in these most economically developed megacities such as Beijing (Chen et al. 2018). Hence, one may interpret the resistance against the rural-urban divide as a locally distinctive characteristic of Chinese rural migrant workers’ anticapitalist struggles, while also recognizing it as an analytically distinct intersectional direction of resistance (Collins 2019).
This section has established that the third SSI qualifies as a work-play of resistance. This SSI occurs continually. None of my informants failed to build weak ties with some customers. This SSI not only yielded resistance against capitalism. It helped Chinese rural migrant workers resist the state-led exploitation and socially established segregation of them.
The Fourth SSI: Developing Sexual Relations with Female Customers
Identifying and Analyzing the Fourth SSI
The fourth SSI that I observed at DG was relevant to cisgender heterosexual male waiters and not female and sexual minority waitpersons. It concerned the development of intimate sexual and/or romantic relations between a male waiter and a female customer. I found that such relations were common among the waiters of DG. I also noticed such relations happening in clubs across China, and they have been reported in some Chinese-language studies (Ma 2006). How such intimate relations constituted an SSI requires a contextual explanation.
Rural migrants in China are not only a socioeconomically disadvantaged group. Since the 1980s, they have been racialized and assigned a low social and cultural status (Han 2010). Rural-urban symbolic boundaries powerfully mark Chinese rural migrants as lowly and dark skinned and Chinese urbanites as suave and pale skinned. Both male and female rural migrants were stereotyped as unhygienic, physically unattractive, and low-quality persons. Furthermore, masculinity in postreform China has been increasingly defined by wealth and political power (Liu 2019). Male rural migrants were seen by urbanites and themselves as seriously lacking masculinity (Choi and Peng 2016). Because of these realities, sex and dating across rural-urban boundaries are reported to be rare (Wang and Nehring 2014). The fourth SSI directly and strongly transgressed these rural-urban symbolic boundaries.
Before working at DG, informant 19 was well aware that rural-urban boundaries precluded his chances of dating urban women. He was surprised that it became possible in the club workplace. He “heard from coworkers about it, still found it incomprehensible” that urban women could accept him as an intimate partner. Informant 8 “gained much confidence from” these intimate relations. Informant 36 felt that these female intimate partners acted as “a rare source of moral support and encouragement for him in the city.” Informant 32 was moved when he heard from an intimate partner that he was “as decent as anybody she had met in the city.” I interpret this instance as evidence for the undermining of racialized boundaries between urban and rural Chinese.
The Fourth SSI’s Resistance Implications
Admittedly, there are some normative complexities and ambiguity in interpreting heterosexual intimate relations. I briefly unpacked them here. I propose to separately treat the evaluation of the undermining of racialized rural-urban boundaries and the evaluation of these heterosexual intimate relations. Undermining racialized rural-urban boundaries is by itself a progressive relational transformation; this is not a contested evaluation. What is contested is the evaluation of heterosexual intimate relations as progressive. An observer may positively evaluate these relations by, for example, emphasizing how they reflect the sexual liberation and agency of women living under a sexually repressive Chinese patriarchy. Another observer may negatively evaluate these relations because these relations are heterosexual and therefore heteronormative and conservative, and the involved waiters are unethical and unprofessional for having sex with clients. This study is not responsible for settling these contested normative evaluations. Moreover, perfect victimhood is not a prerequisite for worker resistance.
The fourth SSI’s resistance implication resembles that of the third SSI. The fourth SSI cultivates new sexual and romantic relations that partly occur outside the dance club. It constructs consecutive durations of short-term intimate relations that incrementally reshape rural migrants’ future relations with urbanites. It undermines existing hegemonic social relations based on racialized rural-urban boundaries and class hierarchy. When the fourth SSI relationally undermines racialized rural-urban boundaries, it generates resistance that is both anticapitalist and intersectional. Like the third SSI, the fourth SSI helps workers resist the Chinese urbanite society’s bottom-up driven oppression of rural migrants on one hand, and the Chinese state’s top-down-driven oppression of rural migrant workers on the other hand.
I found that the fourth SSI occurred recurrently. Because of the secrecy surrounding sexual issues, I was unsure how exactly frequently it occurred. I relied on an indirect measurement to estimate it. I analyzed DG waitstaff’s monthly expenditure patterns and found that most of my male informants reserved between 200 and 300 yuan each month for renting love hotels. Three hundred yuan was enough for two nights’ rent in 2015. Therefore, I estimate that the fourth SSI occurs once or twice per month for each waiter.
Conventional Resistance Triggered by the Four Work-Plays
The four work-plays triggered further worker resistance in DG. Importantly, this resistance was explicitly oppositional, nonrelational, and noncultural: it is clearly interpretable as conventional resistance. I observed three such instances of resistance between July 2014 and June 2015; this section analyzes one of them. This analysis helps reinforce this study’s argument that the four work-plays emphasize resistance rather than consent. It also showcases the variety of ways that work-plays contribute to resistance.
The first instance of conventional resistance involved the waitstaff’s campaigning to cancel DG’s “foreigner-guest policy.” The term foreigner-guests referred to specific non-Chinese customers who were registered by DG as preferred clients and were offered free entrance, free tables, and free drinks. A significant minority of clubs in China set up such a policy to make the club appear cosmopolitan, exotic, and authentically global to local clubbers. My informants disliked foreigner-guests because most of them refused to participate in my informants’ work-games and work-plays. Foreigner-guests refused to do so because, as either Western clubbers or non-Western yet globalized clubbers, they did not enjoy local Chinese work-plays and work-games. For example, they prefer interacting with other clubbers and with Chinese clubbers rather than service workers. Because they did not endorse the work-games, they never paid tips. Because they did not endorse the work-plays, they seldom let waitpersons develop SSIs with them.
Initially, my informants discussed the foreigner-guest problem and brainstormed plans to cancel the policy. They did it during the late-night postwork meals. These discussions took place among different networks among the waitstaff for more than two months. The initiators ensured that they convinced most of the waitstaff. In this process, they also convinced some low-level managers. Then, the initiators and low-level managers negotiated with the midlevel managers. They gave well-formulated reasons, such as testimonies of how the unruly behaviors of foreigner-guests antagonized other customers. Several informants hinted they could no longer stand the foreigner-guests and wished to quit DG. This campaign succeeded, and the foreigner-guest policy was canceled.
The second instance of conventional resistance concerned the trial implementation of a new kind of waitstaff: the “model-waiters.” It was to be staffed by tall and handsome Han Chinese young men employed through modeling agencies. Being the center of the modeling and celebrity industries in China, such men were in good supply in Beijing. Model-waiters specialized in the aesthetic labor of looking good and presenting the club as a stylish institution to customers. An unintended secondary effect is that my informants, or the old waitstaff, would come to exclusively do manual work. The model-waiters would stand near customers and interact with them. The old waitstaff would do the running between the customer tables and the club’s backstage (i.e., the kitchen, the bar, and the inventory room). Although the model-waiters took a cut in the tips, they would not significantly lower the old waitstaff’s overall income. To my informants, the threat of model-waiters was about the second SSI and the associated work-play. Informant 12 was angered when a few model-waiters started working alongside him; he felt that he was “losing something important” in the new working condition. I interpret his anger and feelings as a reaction against being bereft of aspirational capital. Similarly, informant 34 complained,
The model-waiters practically cut off the connection between our customers and us. I did not come here just to bus tables or do other meaningless labor. . . . If the model-waiter arrangement were really implemented, most of us would have quit DG.
Fortunately for my informants, model-waiters were not a widespread arrangement in the industry. DG’s managers were only experimenting with it. When the managers asked the waitstaff to evaluate the model-waiters, my informants tried their best to vilify them. A midlevel manager informed me that the management did not entirely believe the waitstaff’s evaluation. Nonetheless, they still scrapped the model-waiter arrangement because “they sensed the staunch resistance of the waitstaff.”
The third instance was a protest against the general manager, David. David was a Taiwanese who had managerial experience in London’s high-end clubs. In my observation, he was not a particularly bad or exploitive manager. However, my informants disliked him because he strongly suppressed the waitstaff’s local practices. My informants’ work-games and work-plays were a focus of suppression because David thought they looked provincial and would damage DG’s stylish and cosmopolitan image. For example, he tried to minimize the waitstaff’s interactions with customers by educating the waitstaff to serve customers in a socially detached way. I observed that both the initial training program and regular waitstaff training sessions asked my informants to “serve customers as an English butler would do.” The arrangement of model-waiters was also initiated by David.
As numerous minor conflicts accumulated, my informants became convinced that they had to fight to oust David from DG. Like the first instance, my informants organized themselves as a collective and talked to the low-level managers. Unlike the first instance, they additionally organized all varieties of frontline service workers in DG to protest. The salespersons, cashiers, and security personnel were also sympathetic to the waitstaff. To show their determination, the protestors threatened to collectively quit DG if David remained. A midlevel manager explained to me that collective quitting would not fatally hurt DG, but it would be sufficiently powerful to stop DG’s operation for a few weeks. My informants knew that the disruption they could bring would be limited, so they tried diversify their protest action. For example, they initiated a social media campaign to ask all nighttime service workers in Beijing not to join DG. There were already established WeChat (the Chinese equivalent of WhatsApp) groups dedicated to Beijing’s nighttime service workers. Most nighttime service workers in Beijing have joined these groups. DG’s low-level and midlevel managers were receptive to workers’ requests. They also thought David did not understand mainland Chinese society well; a midlevel manager noted that “a cosmopolitan atmosphere did not necessarily maximize revenue for the club or increase customer flow.” Shortly after these protests, David was transferred to another club of the corporation.
Work-plays (and work-games) are inserted via informal and bottom-up ways into the labor process by workers. They are not guaranteed to survive and recur. Given their relational displacement and transgression of capitalist relations, managers likely wish to terminate them. The primary social force that keeps them alive would be the workers themselves. Namely, workers must fight against the management to defend their work-plays of resistance. Such a defense is likely to lead to direct and overt conflicts with the management. The three instances illustrate this chain of resistance and conflict. My informants eagerly defended work-plays. Such resistance assumed the forms of protests, strikes, and negotiation, forms that are conventionally categorized as industrial action. These instances hint that work-plays can be resilient and qualify as informal work routines.
Conclusion
The previous analysis offered an answer to the long-standing and important question of whether there are service work-games of resistance. Adopting Lopez’s (2010) idea of SSIs and demarcating between work-games and work-play, I uncover the reasons why service work-games of resistance have not been found. The gamelike characteristics of work-games are inconsistent with the noninstrumental nature of SSIs. Then I proposed the general term ludic informal work routines to cover both work-games and work-plays and to serve as work-games studies’ general research object.
The search for service work-plays of resistance would better serve the original theoretical goal of sociologists of service work, including McCammon and Griffin (2000) and Lopez (2010), which is to understand whether and how ludic informal service work routines generate resistance. Although some studies have already found worker-customer interactions that resemble SSIs, two further tasks were needed to advance this search. We did not know how recurrent these SSIs were, a crucial examination for ascertaining whether the interactions qualify as work-plays. Additionally, it was unclear how these SSIs would contribute to resistance. My analysis performed these two tasks and supplied useful findings.
I showed that the four SSIs observed at DG were recurrent. The first task’s analyses were relatively straightforward; a minor complication is that these SSIs do not show the kind of regimented regularity found in certain classic factory work-games such as banana time. Despite this, given that some of them occur daily or weekly, it is reasonable to regard the four SSIs as informal routines and work-plays. The second task’s analyses were relatively complex. I adopted the concept of relational resistance to explain how exactly the four work-plays constituted and generated resistance. My explication clarified that relational resistance operated through momentarily displacing existing capitalist relations and constructing noncapitalist relational substructures that overlay established capitalist structures. It also demonstrated that the four work-plays’ relational resistance did not merely target capitalism but was intersectional. Additionally, I illustrated how the four work-plays triggered nonrelational and conventional types of worker resistance. Overall, these analyses yielded strong empirical evidence for confirming the existence of service work-plays of resistance.
DG officially closed its doors in 2022, and it had already become inactive in 2020 because of the coronavirus disease 2019 (COVID-19) lockdown. On the basis of a follow-up round of data collection in 2026, I found that the first, third, and fourth work-plays largely remained intact in Beijing’s clubs, though the waitstaff is confronting slightly less favorable circumstances for them. The major reason is that, between the mid-2010s and 2026, the nightlife market in Beijing and other Chinese cities has been contracting. My informants believe that this was caused by the broad contexts of China’s economic peaking in the late 2010s, the lockdown caused by COVID-19, and citizens’ increasing preference for online sociality. The number of clubs in Beijing’s nightlife district moderately decreased. The Club waitstaff needs to serve more customers and enjoys less free time during work now. Consequently, chances for partying with coworkers modestly decreased, while chances for copartying with customers remained unchanged. The second work-play was undermined by the moderate decline of the Chinese club scene. Dance club managers offer fewer freebies now. Despite this, the second work-play was not entirely blocked; it simply exists in smaller scale. The third work-play still exists. All six informants told me that waitpersons still eagerly develop weak ties with customers. It is difficult to measure whether customers are less receptive to such ties due to economic reasons, however. One of my informants, a manager located in Shenzhen, said that customers are slightly less receptive to these ties compared to the 2010s. The two informants who work as waiters report that the fourth work-play, based on intimate relations with female customers, still exists. One of them even thought that the economic downturn in the 2020s enhanced these intimate relations. Although these changes are limited, they illustrate that work plays can be fragile and unstable. Their survival depends on workers’ proactive efforts.
Given that this is the first study of work-play and one of the first sociological analyses of service workers’ relational resistance, I hesitate to broadly generalize its findings or derive big theoretical conclusions from them. Although service work-plays of resistance exist in China’s clubland, they may (or may not) be rare across the world. The Beijing dance club workplace setting may be more conducive to work-plays of resistance than other workplaces.
The previous analysis’s major limitation is that it has not proven that relational resistance is as impactful for social change as conventional resistance or cultural resistance. For example, although it illustrated that work-play generated a nonnegligible amount of resistance, it did not prove that work-play alone can create ideal work environments or end capitalism. The more relational resistance is recognized as inherently impactful, the more strongly I can argue that the SSIs that this study found are service work-games of resistance rather than service work-games of consent. The proof of relational resistance’s impact involves much more than a single study’s arguments and analyses. The proof and recognition of cultural resistance’s impact have taken a few decades, elaborate theoretical arguments, and numerous empirical studies. Current studies of relational resistance are making similar efforts, though lagging behind the efforts of recognizing cultural resistance. If the symbolic worker actions of cultural resistance qualify as resistance, I do not see how the practical actions of relational resistance cannot qualify as resistance.
Two speculative thoughts on relational resistance can be extrapolated from the previous analysis. I outline them to help facilitate future efforts of exploring relational resistance. The first is that relational resistance seems to be primarily composed of transformative relational practices instead of direct confrontations with (physical and cultural) capitalist structures, though they can indeed trigger such confrontations. My second thought is that relational resistance does not create impact by taking the paths of socialist revolution and state projects, labor electoralism and welfarism, or cultural changes of citizens’ minds. Instead, it seems to do so through the direct action of relational practices to dealienate work and displace capitalist relations of production. In this light, relational resistance is more akin to situationism, anti-work, and the utopian vision of early Marx than the critical cultural theories of Foucault and Gramsci or the late Marx’s Communist vision.
Footnotes
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partly supported by the General Research Fund (241813), University Grants Committee, Hong Kong SAR.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.