Introduction

Every society prevents crime, resolves conflicts and maintains its order through a variety of formal and informal forms of social control. Widely recognized differences in the exercise of social control are practiced in Western societies and Eastern societies (Jiang et al. 2021; Zhang et al. 2017). While the West focuses on the binary control system (formal vs. informal), China is characterized by interdependent and collectivistic orientations fused through a set of practices coproduced by community mediators and formal control agents, which is conceptualized by scholars as a form of semiformal control (Huang 1993a, 1993b; Jiang et al. 2013, 2015, 2019). Previous studies suggest that semiformal social control is prevalent and plays a crucial role in social governance and crime control particularly at the grassroots community level in Chinese society (Huang 1993a, 1993b, 2008, 2019; Jiang et al. 2013, 2015, 2019, 2021, 2022; Messner et al. 2017; Zhang et al. 1996). Research clearly indicates that societies may employ various forms of social control, depending on preconditions found in their historical and cultural development, their political systems, and their moral and ethical foundations. Therefore, investigating how semiformal control is shaped by these social forces in China brings a deeper understanding of the compatibility of formal and informal social control and its variants in non-Western societies. However, there is limited research that has empirically examined the correlates of semiformal control in China, particularly in rural China where classic Confucianism-based control practices are deeply embedded.

The current study aims at filling the void and investigating the correlates of the prevalence and importance of the observed trinary (formal, semiformal and informal) control system in rural China. The prevalence of different forms of control in this study refers to the observed prevalence of government, police (two formal control mechanisms), village committee (a semiformal control mechanism) and the elite gentry (or xiangxian, an informal control mechanism) participating in the community order maintenance activities by survey respondents. The data used in this study includes village (structural) and individual variables, thus the research examines the predictors at the village and individual levels simultaneously.

In the next section, we will first define social control and semiformal social control, along with the definitions of formal and informal social control. We will then review the predictors of the prevalence and importance of the observed forms of social control. The Methods section follows, and then we interpret the research findings. Finally, our study will conclude with a discussion regarding those predictors involved with semiformal control and the other two forms of social control.

Social control and semiformal social control

It is commonly recognized that Edward A. Ross is the main innovator of the concept of social control. He defines social control as “the molding of the individual’s feelings and desires to suit the needs of the group” (1896: 518). According to Ross, the primary function of social control is to maintain and contribute to social order. It is a mechanism to counter deviance, facilitate social interaction and underscore restoration (Trevino 2019). Given Ross’s legacy, Talcott Parsons and other scholars propose similar definitions of social control (Chriss and James 2013; Meier and Robert 2019). A conceptual definition of social control in this study is that social control is any attempt to counter deviance and crime and mediate disputes or resolve conflicts. All these attempts are community order maintenance activities. Countering deviance and crime includes preventing and fighting deviance and crime. Dispute mediation is a process where a third party intervenes between conflicting parties to resolve their differences (McCorkle and Reese, 2015). The third party in rural China can be village committees, local governmental agencies including the police, and xiangxian. Countering deviance and crime and mediating disputes are two important approaches to maintain social order.

Research has repeatedly focused on examining how various forms of neighborhood control, especially informal control, operate to prevent or reduce the prevalence of social disorder (Zhang et al. 2017). Parochial informal control sometimes interacts with public levels of control to give rise to a set of semiformal practices carried out by neighborhood residents and formal control agents (Carr 2003). Wojkowska (2006) defines the informal system as those that function outside formal legal structures, such as the police, prosecution, courts, and penal systems. As an alternative approach to resolving conflict outside the formal legal framework, the informal system allows for a larger range of emotional expression and involves real-life mediations among individuals. Guided by the theory of restorative justice and the ideals of restorative communication, the system includes various types of structures and methods for resolving conflict, such as traditional and indigenous structure—religious courts, and semiformal institutions—community courts (Rani 2019). For example, reparative board meetings, typically comprising local residents and a coordinator from the Department of Corrections, adhere to semiformal guidelines in addressing offender issues, reflecting a semiformal control approach (Dzur and Wertheimer 2002).

This semiformal control is especially evident in Chinese society. Different from the West, where restorative justice functions more as a complement to the formal system, in China, the use of semiformal control is integrated as an essential part of community governance. This integration is evident through various local organizations, which, while being marked by government affiliation, retain limited autonomy and involve both formal and informal control features. Research in the past few decades has documented the prevalence of semiformal social control performed by various semiformal organizations in China (Huang 1993a, 1993b, 2008, 2019; Jiang et al. 2013, 2015, 2019, 2021; Messner et al. 2017; Zhang et al. 1996, 2007). Early research spanning Qing dynasty China into New China by historian Philip C. C. Huang (1993a) introduced the term “third realm” to describe how effective semiformal control to maintain social order in Qing dynasty was forged between the court system and societal mediation through xiangbao, an unsalaried quasi-official nominated by the local community and confirmed by the state. The semiformal process involving interactions between formal and informal control and quasi-official intermediaries almost became a standardized part of the court administration to resolve the “minor matters” (Huang, 1993a). In the traditional Chinese legal system, “(P)ersisting features include the mode of practical moralism in legal thinking, the combination of mediatory and adjudicatory justice under a single system, and basic characteristics of court adjudication and mediation” (2010: 214). In their study of social control in contemporary urban China, Messner et al. (2017) found that indicators of two important components of semiformal or “semipublic” control, which they refer to as tiao-jie (i.e., mediating and solving disputes and conflicts among residents or between family members) and bang-jiao (i.e., assistance, guidance, and direction to residents who have committed minor offenses), are negatively associated with reported property victimizations of members of the household that occurred within the neighborhood.

Within the duality of the Chinese culture and authoritarian institutions, semiformal organizations and their practices in China demonstrate features distinctive from those of Western societies. These features have been well exemplified and systematically elaborated in a variety of studies of social and crime control in Chinese communities conducted by Jiang and his colleagues (e.g., Jiang et al. 2007, 2013, 2015). According to Jiang et al. (2015), “semiformal organizations integrate formal and informal organizational features; they are grassroots organizations but are marked by the government; they are nongovernmental organizations but have official power and responsibilities; and they are a bridge between the state and society” (p. 290). Resident committees (juweihui) in urban areas and village committees (cunweihui) in rural areas are the most prevalent forms of semiformal organizations in Chinese communities bridging the state and society. Community committees including village committees are self-governing organizations encompassing local resident members who implement policies from the central government and provide services to local residents, such as settling disputes in the community, working with the local police department to maintain public order, helping former offenders reenter society, and assisting household registration (Shou, 2015; Tsai, 2002; Yan and Gao, 2007; Mittelstaedt, 2022). Community committees also play an important role in restorative justice in China, such as people’s mediation in civil conflicts (Zhang and Xia, 2021) or criminal reconciliation (or forgiving conferences) in criminal cases (Jiang and Chen, 2023). However, the community committees are by no means voluntary associations but instead operate as government-supported grassroots units “required and expected to assist street officers to serve government roles” (Jiang et al. 2013, p.210).

Semiformal control in this study refers to controlling action implemented by semiformal organizations using sentiment, reason, and law. In contrast, formal control refers to a control action “imposed by law and/or implemented by official controlling organizations”; informal control refers to a control action “carried out by unofficial controlling groups based on moral rules” (Jiang et al. 2007, p.262). Understanding different forms of control in a given society can provide valuable insights into the mechanisms of social order and the ways in which communities maintain stability. Recently, Jiang et al. (2022) empirically compared and contrasted the prevalence and importance of a trinary control system (i.e., formal, semiformal, and informal) in community order maintenance in contemporary rural China. Their study found that control from semiformal organizations, that is, village committees, appears to be the most common mechanism for mediating within-village disputes. Village committees, alongside government officials also participate more frequently in mediating beyond-village disputes than do the police and xiangxian (village informal control). Their findings suggest that social order in rural China is largely shaped by the integrated control built upon the trinary control system, in which the village committees coordinate with formal and informal institutions and maintain a level of autonomy to resolve disputes, combat crime, mobilize resources, and establish community solidarity through collective efficacy.

Despite the significance of this line of recent research on community control and crime prevention, questions remain, including how community structural and individual conditions attenuate or enhance the prevalence and importance of different forms of social control in the villages of rural China. Recently, Jiang et al. (2021) examined individual-level correlates of the views on social control mechanisms involving both urban and rural areas in China. Yet, the study exclusively focused on individual-level correlates, and it did not examine social control in rural China separately. The present study, therefore, compares the extent to which community characteristics and individual factors explain the prevalence and importance of semiformal control, formal control, and informal control. In the following section, we discuss the structural and individual measures employed by this study and their potential link to social control.

Structural factors contributing to social control

A large body of literature exists discussing neighborhood capacity to engage in social control and the importance of maintaining a strong social organization to explain variations in the level of local crime and disorder (e.g., Mazerolle et al. 2010; Morenoff et al. 2001; Sampson et al. 1997; Sampson 2012). Alongside the proliferation of research that addresses the effects of social control, growing research attention focuses on the sources of social control in neighborhoods (e.g., Browning et al. 2017; Burchfield 2009; Silver and Miller 2004; Sampson et al. 2002; Warner 2007). This neighborhood level line of research has found that neighborhood structural characteristics, such as higher levels of disadvantage and racial heterogeneity are related to social ties and cohesion, which can affect a neighborhood’s ability to regulate local behavior (Warner 2014)Footnote 1. The existing research, however, predominantly focuses on the sources of informal control in Western societies. Studies rarely establish a connection between the structural conditions and semiformal control, prevalently used to regulate social order in Chinese society. Our study advances the previous research by incorporating structural features, including economic disadvantage and racial heterogeneity, into an examination of the sources of semiformal, formal, and informal control in rural China.

Economic disadvantage

Concentrated poverty, or economic disadvantage, deprives neighborhood resources and the capacity to form strong informal social control as poverty deteriorates social cohesion and trust, impedes and limits communication, and weakens the perception of shared norms within the neighborhood (Kubrin and Weitzer 2003). Neighborhoods with high concentrations of poverty in the United States exhibit weak informal social control over youth (Silver and Miller 2004). Results of neighborhood level research regarding the connection between economic disadvantage and social control in China are somewhat inconsistent. In their investigation of the sources of social control in the city of Guangzhou, China, Jiang et al. (2013) found that increases in economic disadvantages of communities are associated with decreases in residents’ obedience to the law. This association was indicated by percentage unemployed, percentage below the average economic status, and percentage of non-homeownership. Based on the same sample, Jiang and his associates (2013a, 2013b) further examined informal control, formal control, and semiformal control. Their findings demonstrated significant effects of community poverty on semiformal control measured by residents’ reported importance of neighborhood’s mediation committee on maintaining community order, but not on formal control and informal control. Similarly, in Messner et al.’s (2017) study in Tianjin, China, community poverty was not associated with the reported neighborhood watch activities as an indicator of informal control. Messner et al. (2017) attributed the absence of an association to the incompatibility of applying Western conceptualization of social control in an Eastern society.

Previous studies highlighted the importance of adopting measures of “‘neighboring’ that might be conceptualized as reflecting social cohesion, especially as manifested in a collectivistic society, such as China” (Messner et al. 2017, p. 563), as well as specifying structural impacts on different forms of social control. Therefore, in addition to the measure of the percentage of poor families in a village community, we include the level of village collective revenue that specifically measures economic indices in rural China. In most of China’s rural areas, the livelihoods of people are closely tied to their land. The village collective is a common feature of land and income allocation that solidifies the village community as a collective through organizing agricultural production, land management and resources within these rural villages and distributing financial benefits to rural households based on labor inputs of household farmers (Wang and Tan 2020). Status and income differentiation will affect the community mediation system which best suits villagers of roughly equivalent economic power. Revenue of village collectives has been the source of funds for public welfare, services, and subsidies to others (Oi 1999). In poor villages where village committees control little, or even no, collective wealth and resources, the villagers’ stake in village collectives is low (Hu 2005). This may further weaken the informal and semiformal control in villages that heavily rely on villager support and participation. Thus, we expect that higher percentages of poor families and lower levels of village collective revenue are associated with lower levels of informal control and semiformal control in a village community.

Ethnic heterogeneity

Chinese rural village residents relocate permanently far less frequently than urban residents. Residential proximity, rather than mobility, characterizes rural areas where traditional culture, customs and hierarchy endures. Though different ethnic groups in a village may, to a certain degree, have differences in culture and customs, they have long shared a collective ideology with many rules and customs. Their differences in terms of culture may not be as visible as in Western countries. The predominance of homophilic social ties may not exist. Thus, we hypothesized that ethnic heterogeneity does not affect the prevalence and importance of informal and other forms of social control.

Individual factors contributing to social control

Besides the influence of structural factors, social control efforts, especially informal social control, are the product of actions by individuals (Coleman 1990). Understanding the formation of social control should partially be rooted in individual-level processes through which individuals make the decision to take social control actions. Accordingly, in addition to structural factors, this study includes several individual factors identified as important associates of social control in prior research (Jiang et al. 2010, 2014, 2021; Lambert et al. 2012).

Village cadre

Village cadres refer to members of a village committee nominated and elected by village fellows and approved by a superior authority or township government. Unlike grassroots bureaucrats in many countries, village cadres in China are not employees of a single agency assigned a limited number of duties. Instead, they are “bureaucrats-cum-politicians answerable” for a village’s all-around governance and development (O’Brien and Li 1999). Cadres are responsible for keeping social order and providing services to village residents. They are also liaisons between township government (or government agencies) and villagers. Village cadres are more likely to observe the participation of village committees, government and police, as well as the xiangxian or kin-kind community leaders in social order maintenance activities. This study hypothesized that compared to regular villagers, a cadre is more likely to observe the participation of a village committee, township government, police, and xiangxian in community order maintenance activities and rank the importance of each form of social control mechanisms higher.

Political affiliation

The Chinese Communist Party (CCP) is the leading party in China. The leading position of CCP has been institutionalized and written into the party constitution exercising overall leadership covering all areas of endeavor in every part of the country (Trevaskes 2018). The appointment of members throughout the government, other public institutions, nonpublic organizations, and other positions of prominence and power at many levels in China are restricted to party members (deLisle 2022). CCP members are expected to actively participate in or lead village activities including community order maintenance. Thus, this study hypothesized that the party members are more likely than the non-party villagers to witness the participation of different forms of social control in community order maintenance activities and in turn develop a more favorable perception of the importance of those control mechanisms.

Gender

In the West, researchers have observed that gender differences in attitudes towards control, with women often being associated with an empathic and caring attitude towards criminal justice issues, while men tend to be more impartial and justice-oriented (Gilligan 1982). While women in modernized China gained higher education and economic independence, patriarchal tradition, collectivism, and the one-party political system still encourage gender similarities in thinking (Lambert et al. 2007). Previous studies on the association between gender and formal/informal social control in China have yielded mixed results. Although a few studies reported a relationship between gender and formal/informal social control (e.g., Jiang et al. 2010), most studies did not find any associations (Jiang et al. 2013, 2021, 2013, 2007; Wu et al. 2011). Based on past studies in China, the current study hypothesized that gender is not related to the reported prevalence and importance of formal, semiformal and informal social control in community order maintenance.

Age

Younger people are commonly believed to value social control less than older people as they are reported as being freer from constraints and more self-supported and self-sufficient. Among the studies quantitatively examining whether age is associated with views of social control in China, four studies (Jiang et al. 2013, 2014, 2010, 2021) revealed no relationship existed between age and the views of informal social control. One study (Jiang et al. 2010) reported a positive relationship between them. Six studies found that age is not associated with the views of formal social control or formal control organizations such as police or other government agencies (Jiang et al. 2013, 2013, 2010, 2014; Wu et al. 2011; Jiang et al. 2012). Two studies found a correlation between age and views of formal social control or formal control organizations (Jiang et al. 2021; Sun et al. 2017). Only two studies examined the relationship between age and the views of semiformal social control. Jiang et al. (2013) found that age is negatively associated with semiformal control while Jiang et al. (2021) reported no correlation exists between the two variables. Based on the idea of youth valuing social control less, this study hypothesized that age is positively related to the reported prevalence and importance of all forms of social control.

Education

Compared to gender and age, education is not commonly used as a predictor of views of social control in China, with mixed results connecting education to the views of different forms of social control. Highly educated people tend to exhibit a dual inclination, valuing formalization or the rule of law while also favoring free choice and less formal control. Therefore, studies examining the relationship between education and views on social control in China have produced mixed results. Some found no association between education level and the reported importance of informal control (Jiang et al. 2010), while some reported that more educated people place greater value on informal groups but less on formal organizations (Jiang et al. 2009). Giving the reasons above and the empirical findings, this study invoked no specific hypothesis regarding the connection between education and views of different forms of social control but included education as a control variable.

Data and methods

Data

ThisFootnote 2 study is based on the 2021 survey project organized by a highly regarded, university research center in Hubei Province, China that focuses on rural China. The center has conducted village surveys nationwide every year since 2006, except in 2020. The 2021 survey was conducted at both the village and individual levels. Both levels of questionnaires included a basic data section collecting socioeconomic data and eight theme-specific sections pertaining to personal observations of community activities and opinions. Village governance is one of the eight themes where data showing the prevalence and importance of forms of social control were extracted.

The center’s previous surveys were used to develop the basic data section questions. The village governance questions were selected based on the literature review by the survey organizer who has published papers on social governance. The survey organizer conducted previous survey research in China, has conducted graduate student interviewer training on survey methods and has collaborated with international scholars in publishing their scholarly papers. The final version of the village governance questionnaires was developed through pretests with scholars, villagers, and some interviewers.

The village setting and resulting samples in this study are closely associated with the interviewer selection. Interviewers selected in this study must meet the following criteria. First, candidates were graduate students who had completed a research methods course or had previous experience administering surveys. Second, candidates had relatives living in the countryside during Chinese New Year (February 12, 2021) so that they could administer the survey during their holiday break. The Chinese countryside is still an acquaintance society. With a rural background and acquaintance with their parents’ or relatives’ village, interviewers were able to access village leaders and fellow villagers. Importantly, village kinship status was exceedingly helpful to successfully disseminate the surveys on social control. China has many local spoken dialects even though written Chinese is the same. People with different dialects may not readily understand each other. Villages may have very different customs and norms as well. Sharing the same dialects and culture is essential for relating to respondents during the survey. It is also helpful for an interviewer to check if a respondent provided accurate information.

Besides the quality interviewer selection, the village sample selection in this study also considered its geographical breadth, economic development level, and types of villages (whether a village is near or within an urban area). Samples are from 25 of 31 provinces (or equivalents), representatives of eastern, western, northern, southern and central China. Based on 2020 GDP per capita, the 25 selected provinces included the richest ones such as Jiangsu, Fujian, Zhejiang, GuangdongFootnote 3 and the poorest ones such as Gansu, Heilongjiang, Guangxi and Guizhou. Within these selected provinces, the survey organizer also considered the geographical breath of counties, selecting one village per county. Accordingly, 164 villages in this study were from 164 counties. When selecting the villages, the research team generally knew the economic statuses of each county based on 15 years of previous nationwide village surveys and national statistics. Thus, the survey organizer selected the villages representing different economic status. The sample results from this study confirmed that 164 villages include a variety of economic development levels. For example, collective revenue, an indicator representing the economic status, varied from 0 to 12,019 yuan per resident among the selected villages. Since the distance of a village to the nearest urban area may affect the village’s economic development, mobility and interaction among villagers and community safety, the village sample was chosen to balance urban proximity. The selected villages included villages that are not within or near an urban area (approximately 90% of the sampled villages) and villages that are within or near an urban area (10%).

The interviewers were trained by experienced researchers from the Center. To help improve the survey quality, all the interviewers and the survey organizer joined a survey group on WeChat, a popular Chinese social media app, for communication until the completion of the survey. The surveys were conducted during the students’ winter break in 2021—between January and February. The data collection method was face-to-face survey interviews. The in-person interview was important to improve survey quality because many respondents in the countryside have little formal education. Each interviewer was expected to interview 15 villagers for the individual questionnaires and one village leader for the village questionnaire in his or her village (Jiang et al. 2022). A total of 164 villages (or village leaders) were included in the sample interviewed. A total of 2460 individual surveys were distributed and 2343 people completed questionnaires at the individual level, a response rate of 95 percent.

Dependent variables

Dependent variables in this study are the prevalence and importance of four types of organizations or groups in 10 village order maintenance activities. As for the prevalence, the question was “Overall, which organizations or groups participated most frequently in the following 10 village order maintenance activities? (The most = 4, the second most = 3, the third most = 2, and the least = 1)”. This is a matrix question. Organizations and groups include xiang or township government and upper-level government, paichusuo (or community police), village committee, and xiangxian (or the elite or respected people, not including current village leaders). The 10 activities are (1) family dispute mediation, (2) house property dispute mediation, (3) villager dispute mediation, (4) villager vs. village committee dispute mediation, (5) villager vs. village-owned economy dispute mediation, (6) villager vs. government dispute mediation, (7) villager shangfang (appealing to upper government) mediation, (8) quntixing shijian (mass incident) mediation, (9) land expropriation and housing demolition dispute mediation, and (10) social order maintenance such as preventing and controlling crime, violence, and drug abuse.

As for importance, each of the four mechanisms above was evaluated in each of the above 10 activities. The answer categories for each question were very important (= 5), important (4), not sure (= 3), not important (= 2), and not important at all (= 1). A complete list of these measures is found in Tables 2 and 3.

Independent variables

Village variables include village collective revenue, poverty, and ethnic heterogeneity. Collective revenue was measured by village collective revenue per capita. It includes money from rent, village-managed enterprises, donations, government subsidies, and revenue balances transferred from the previous year. Poverty was measured by the proportion of poor families in a village, including dibaohu or households whose mean income is below the poverty line and wubaohu or households or individuals who are incapable of supporting themselves such as the elderly, disabled, or orphans. Ethnic heterogeneity was measured as whether a village has one ethnic group (= 1) or more than one ethnic groups (= 0). Due to the predominance of the Han ethnicity (about 93% nationwide) in China, it was not easy to collect data including ethnic diversity in Chinese studies. Fortunately, our survey data included ethnically diverse villages in 8%Footnote 4 of the 164 villages surveyed, allowing us to investigate the connections between ethnic heterogeneity and forms of social control.

Individual level variables included cadre, party affiliation, gender, age and education. Cadre measures whether a respondent was the village cadre (yes = 1, no = 0) while party affiliation was measured by the communist party member (yes = 1, no = 0). Gender was measured as male (= 1) and female (= 0). Age and education were measured by years, respectively. As a key level of inquiry, the prevalence of the four forms of control mechanisms in each village order maintenance activity was also viewed as an independent predictor of the importance of the four forms of control organizations or groups in each village order maintenance activity.

Analytical strategy

Multi-level models are employed to investigate the village- and individual-level predictors of the prevalence and importance of the four control mechanisms. The study provides regression modeling for the prevalence and importance of the four control mechanisms in each of the ten village order maintenance activities to investigate the predictors in a specific activity. Similarly, the study includes the regression model for the prevalence and importance of the four control mechanisms in overall of the 10 village order maintenance activities to investigate the predictors in general. The advantage of the regression analysis for each village order maintenance activity is to reveal detailed information on the prevalence and importance while its disadvantage is a lack of content validity in measuring village order maintenance. In contrast, the regression analysis for the prevalence and importance of the four control mechanisms in the overall 10 village order maintenance activities can increase the content validity but may mask specific relationship between predictors and the prevalence and importance in a specific village order maintenance activity.

Results reported in Tables 2 and 3 were from the multi-level linear regression analysis of the dependent variables. Since individual mediation items are ordinal variables, using the ordinary linear regression for estimates is not as robust as using the ordinal regression. Thus, this study used the multi-level generalized linear regression for robust checks. We compared the results from ordinal regression models with those from the ordinary linear regression models, only a few results (regression coefficients) are changed from statistically significant to statistically insignificant. Giving the robust checks, we decided to report results in Tables 2 and 3 that are statistically significant from the ordinary linear regression analyses and the ordinal regression analyses. This approach allowed us to have both robust estimates and simple interpretation of the results. Since we had 70 individual mediation-based dependent variables that were ordinal, simple interpretation or understanding of the results was important.

Findings

Table 1 presents the descriptive statistics of independent variables at the village and individual levels. The average collective revenue per capital varied from 0 to 12,019 Chinese yuan ($1 dollar was about 6.7 yuan at the survey time). The mean percentage of poor families was 10 with a range of 0 to 57. Approximately 8 percent of villages had more than one ethnic group. At the individual level, the mean age was 55.5 with a range of 18 to 98. Female respondents were 35 percent of the sample. The mean education was 7.1 years with a range of 0 to 19. Approximately 6 percent of the respondents were cadre while 9 percent were Communist Party members.

Table 1 Descriptive statistics of independent variables.
Full size table

Table 2 presents the results from the multi-level regression of the prevalence of four control mechanisms in each of the 10 village order maintenance activities and overall activities. Only statistically significant coefficients are included in the table, and the statistically insignificant coefficients are suppressed to facilitate the readability of the table.

Table 2 Two-level predictors of prevalence of four control mechanisms.
Full size table

A visual inspection of the results in Table 2 shows that at the village level, only collective revenue is a statistically significant predictor of the participation level of police and village committee in some of the 10 activities. Furthermore, for villages that had higher collective revenue, their village committee had a higher level of participation in villager vs. village-owned economic dispute mediation, villager vs. government dispute mediation, villager shangfang (appealing to upper government) mediation, quntixing shijian (mass incident) mediation, and land expropriation and housing demolition dispute mediation. In contrast, for villages with higher collective revenue, police were less likely to participate in seven of the 10 activities. These are represented in items 2–7 and 10 in the table: house property dispute mediation, villager dispute mediation, villager vs village committee dispute mediation, villager versus village-owned economy dispute mediation, villager versus government dispute mediation, villager appealing to upper government mediation, and land expropriation and housing demolition dispute mediation. Results for the overall participation in the 10 activities also show that the only statistically significant village-level predictor was collective revenue. Furthermore, village collective revenue was positively associated with the prevalence of village committee in overall community order maintenance action, while it was negatively associated with the participation level of police.

Since the village committee is a semiformal control mechanism, the findings above reveal a positive association between village collective revenue and semiformal control. Police is a formal control mechanism. Thus, the findings above imply a negative relationship between collective and formal control. In addition, village collective revenue is not associated with either xiangxian (a measure of informal control) or the local government (a measure of formal control). In sum, village collective revenue predicts formal, informal and semiformal control differently.

At the individual level, cadre, gender, and age were each statistically significant for some of the 10 activities. Compared to regular villagers, cadres were more likely to observe that the village committee had the highest level of participation in land expropriation and house demolition dispute mediation. Cadres were less likely to observe xiangxian’s participation in village vs. government dispute mediation. However, when all the 10 activities were considered together, cadre was no longer a significant predictor of the prevalence of any control mechanisms.

Males were more likely than females to observe the highest levels of government participation in villagers appealing to upper government mediation, land expropriation and housing demolition dispute mediation, and village safety maintenance (i.e. preventing and controlling crime, violence, and drug abuse). However, males reported a lower level of police participation to address family disputes. Overall, males were more likely than females to observe the highest level of participation of government participation in village order maintenance.

Finally, older people were more likely to report that police had higher participation levels in family dispute mediation and house property dispute mediation, but less likely to observe that the government had higher participation levels in villager vs. village-owned economic dispute mediation. When all 10 items were combined, age was a significant predictor of the participation level of police.

Table 3 presents the results from the multi-level regression of the importance of four control mechanisms in each of the 10 village order maintenance activities and overall activities. As in Table 2, only statistically significant coefficients were included in the table, and the statistically insignificant coefficients were suppressed to facilitate the readability of the table.

Table 3 Two-level predictors of importance of four control mechanisms.
Full size table

All the independent variables in Table 2 are included in Table 3 with an addition of the prevalence of four control mechanisms in 10 community order maintenance activities. It is the same as in predicting the prevalence, village collective revenue was the only significant structural predictor of the outcomes measuring the importance of four types of social control mechanisms. To be more specific, village collective revenue was positively associated with the importance ranking of village committee in activities 7–10. Village committees from a village with higher collective revenue were more likely to play an important role in mediating villagers appealing to upper government, mass incidents, land expropriation and housing demolition disputes, and social order maintenance such as preventing and controlling crime, violence, and drug abuse. However, village collective revenue was not a significant predictor of the overall activities. Like the findings of the prevalence, the findings of the importance reveal that village collective revenue predicts the importance of formal, informal and semiformal control differently. That is, the increase of village collective revenue is correlated with the higher importance rank of semiformal control but does not predict the importance of formal and informal control. Furthermore, the findings of the importance indicate that the correlation of collective revenue with the importance of semiformal control varied for community order activities.

Individual level variables except education were significant predictors for the importance of the four control mechanisms in particular mediations. As expected, prevalence was a significant predictor in all the regression models. That is, for all four control mechanisms, its prevalence was positively related to its importance ranking in each of the 10 village order maintenance activities as well as overall activity scale. In other words, the observed level of participation in maintaining community order activities affects the participants’ ranking of all forms of control—the formal, informal and semiformal. Cadres were more likely than regular villagers to believe that the village committee played a most important role in villagers appealing to upper government mediation and village safety maintenance or crime control. Overall, being a cadre was positively associated with a higher rank for a village committee in community order maintenance. Cadres were also more likely to rank the government higher in villager dispute mediation, mass incident mediation, and land expropriation and housing demolition dispute mediation as well as over activity index. In addition, party affiliation, gender and age were statistically significant predictors of the importance of various control mechanisms in different activities. However, none of these variables were significant predictors of the 10 combined activities.

Discussion and conclusion

Criminological research with a Western perspective has constructed the nature of formal and informal social institutions and has stressed their importance in community governance (Kubrin and Mioduszewski 2019). However, when it comes to research on social control in Chinese society, earlier studies have revealed the prevalence of a trinary control system, with a third type of control, semiformal control, interacting with formal and informal control to maintain social order (e.g., Huang 1993a, 1993b, 2010; Jiang et al. 2015, 2019; Zhang et al. 2007). In addition to formal and informal control, semiformal social control exhibits a negative association with crime and victimization that occurred in the urban neighborhoods and communities in China (Jiang et al. 2013; Messner et al. 2017). This study investigates the structural and individual characteristics of village communities that predict the three types of normative controls in rural China. Our findings show that both individual and structural factors were predictors of the prevalence of the observed controls exerted by village committees (a measure of semiformal control) and police in maintaining community social order. Village collective revenue was negatively associated with the prevalence of police and positively related to the prevalence of community committee in maintaining village order. The reported prevalence of the four control mechanisms (community committee, police, local government, and xiangxian) were significant predictors of their importance in community order maintenance. As expected, cadres, given their readiness to be collectively mobilized, were more likely than average villagers to value the village committee and the government in community order maintenance activities.

The findings of this study have several implications for research on neighborhood/community contexts and social control. First, the findings add empirical evidence to thinking about the application of Western concepts in Chinese rural society. The concepts of formal and informal social control have been developed and almost exclusively tested in the U.S. and other Western countries. However, these concepts fall short when generalizing in non-Western contexts. As Sampson noted, application of control theory in neighborhoods in a variety of societies is necessary “if we are to make further progress in understanding the generalizability of the link between community social mechanisms and crime rates” (Sampson 2006, p. 162). Our study suggests that measures of community structural characteristics (i.e., poverty and racial/ethnic heterogeneity) may not be applicable to explaining the formation of informal control in rural China, possibly due to the unique structural features of Chinese villages. Regarding economic disadvantages, although economic situations remain difficult, compared to urban dwellers, inequality among fellow villagers is not notably high. Economic disadvantage measured by the percentage of poor families at the village level did not explain the prevalence and importance of the observed informal control in a village mediation or crime control activity. However, economic disadvantage measured by village collective revenue did explain the prevalence and importance of the observed semiformal control. As for ethnic heterogeneity, China is dominated by the Han ethnicity. Even if different ethnic groups live together in villages, usually it is multi-generational. In addition, Chinese cultural and political systems emphasize collective unity and harmony, so different ethnic groups in a village share much more commonality than diversity. Social ties and informal control in a village with mixed ethnic groups remain strong and societal pressure reinforces order maintenance. It is worth noting that the measure of informal control in this study differs from measures commonly used in the West.Footnote 5

Second, the findings highlight the important role of one structural feature that reflects a key characteristic of rural Chinese society—a village’s collective revenue—in the community’s capacity for providing semiformal control. The economic condition of a community captured by the measure of village collective revenue is positively associated with the prevalence the observed semiformal control in most community order maintenance activities and the importance of the observed semiformal control in four community order maintenance activities. Collective revenue belongs to a village and is managed by the village committee for the village’s common good. It can be used to hire people, purchase equipment, build up facilities, and compensate for loss, suffering, or injuries. For example, collective revenue can be used to hire safety guards or auxiliary police officers to patrol a village and prevent deviant and criminal behavior. It can also be used to install streetlights, surveillance cameras and purchase patrol bicycles or cars, resulting in a safer community. This usage of collective revenue may explain why villages (a semiformal control mechanism) were observed as the most frequent participants in community safety maintenance activities and a significant predictor of the importance of these activities.

Prior to the abolition of the agricultural tax in 2006, the tax was one of the significant factors leading to conflicts between villagers, village committee, and the local government. It was also a significant predictor of mass incidents and shangfang. Collective revenue was used to help farmers pay the agricultural tax, mitigating peasants’ burdens to resolve the conflicts (Michelson 2007). Today, as China’s urbanization gains speeds, land expropriation and housing demolition in rural areas often lead to disputes between villagers, village committee and local government as well as mass incidents and shangfang (Chen 2016; Peng 2023). Two major reasons behind the disputes, mass incidents and shangfang are compensation for land expropriation, housing demolition and farmers’ employment. Although the upper-level government created policies for compensation, village committees may want to obtain part of the compensation, which may result in disputes between village committee and villagers. Collective revenue plays an important role here to mediate these disputes. Villages with more collective revenue are likely to take less or no compensation. More importantly, villages with more collective revenue are more likely to have successful enterprises, which are used to hire those farmers affected by land expropriation and housing demolition. For these reasons, collective revenue can be frequently used and was actually used, based on the field observations by the research team, to mediate and resolve disputes between villagers, village committee and local government as well as mass incidents and shangfang (Zheng 2019; Guan 2019; Wang 2023).

Third, predictors for each individual activity may differ from those for the whole. Findings from this study show that some predictors of the prevalence and importance of a control mechanism are statistically significant in both individual community order maintenance activities and overall activity indexes while some predictors of the prevalence and importance of a control mechanism are statistically significant in individual community order maintenance activities only. For example, collective revenue is a significant predictor of the importance of the village committee in four of the 10 activities (items 7–10) but is not a significant predictor of the importance of the village committee in the composite index of the 10 activities. The findings suggest that when examining the correlates of the prevalence and importance of four control mechanisms in community order maintenance, we cannot assume that whenever a predictor is significant in an overall index or measure, it must be significant at each of the individual items that compose the index. To check whether a variable is a significant predictor of a control mechanism in individual community order maintenance activities and the overall index of the activities, researchers can best ascertain their significance by examining them separately. The difference in the impact of collective revenue on individual activities and on the overall index of community order maintenance activities highlights the need for a more detailed analysis of social control mechanisms. Such analysis recognizes the complexity of social dynamics in rural Chinese villages, suggesting that the efficacy of control mechanisms cannot be uniformly predicted across different types of community order maintenance activities.

While our findings shed light on theoretical implications on social order maintenance in rural China, our study is not without limitations. First, although our sample was selected nationwide, it was not randomly selected. Thus, we cannot statistically generalize our findings to all Chinese villages. Second, this study focuses on rural areas only. Findings from this study cannot be generalized to cities in China.

Nonetheless, this study is foundational, perhaps unique, in exploring the predictors of the prevalence of village committee, government, police, and xiangxian in maintaining community order. Although Jiang et al. (2021) investigated the predictors of the importance of different forms of social control, this study expanded Jiang et al.’s study in several ways. This study examined not only the individual level predictors [as Jiang et al. (2021) did] but also group (village) level predictors. One of community activities in Jiang et al.’s study (2021)—checkpoint inspection—was directly related to order maintenance, while the current study incorporated 10 village order maintenance activities. Thus, this study amounts to a more comprehensive and reliable approach to studying predictors of community social order maintenance. Jiang et al.’s study (2021) was based on the data collected January–April 2020 during the occurrence of the COVID-19 lockdown in China. The survey data used in the current study was collected in 2021 when the COVID threat was significantly reduced in rural China so that the situation during the survey time was similar to normal times in the countryside. In this regard, the findings from the current study are more likely to express the representativeness of control within Chinese villages. This study found that both individual and village factors were the correlates of the prevalence of village committee and police in maintaining community social order. Village factor—collective revenue—predicted the importance ranking of the village committee and police via the prevalence of the two forms of control in village order maintenance. Thus, it can be expected that relative to regular villagers, cadres were more likely to value village committee and government in community order maintenance activities. The observed prevalence of four control mechanisms was a significant predictor of their respective importance in community order maintenance.

These findings have implications for decision makers in China. Since a village’s collective revenue is positively related to its participation in community order maintenance and in turn to its importance ranking, village leaders should seek rural collective economic development projects in order to support village participation in mediating and resolving disputes between villagers, village, local government and companies. For example, developing village enterprises can hire farmers who are displaced from agricultural work, reducing conflicts, mass incidents, and shangfang activities resulting from the expropriation of their farmland.

In addition, participation level is strongly and positively associated with importance ranking for all four control mechanisms in each of the 10 village order maintenance activities and the combined activities. Although the Chinese reliance on rural village committees places their importance at the center of this semiformal control strategy to maintain collective order, all forms of organizations including government, police and village committee should work collaboratively and increase their actual and reported participation in activities to improve their community standing and efficacy at the village level.