This survey of the literature explores the relationship between religion, self-control, social control, and criminal behavior. This review reveals an inverse relationship between religion and criminal behavior and unmasks the potential spuriousness present in the religion-crime nexus.
3.1. Religion and Self-Control
In examining self-control, researchers have typically examined the ways in which attitudes, beliefs, ideologies, and values, which are internalized, have the ability to influence the behavior of the individual. Subsequently, in examining criminal activity, members of society who have incorporated a set of attitudes, beliefs, ideologies, and values that condemn criminal activity will be less likely to engage in such activity. Religion represents one example of a system that incorporates these elements.
Advanced by Gottfredson and Hirschi, self-control theory states that individuals possessing high levels of self-control, conceptualized as “the differential tendency of people to avoid criminal acts whatever the circumstances […]” (
Akers and Sellers 2013, p. 122), are less likely to engage in criminal/deviant behavior. Comparatively, individuals possessing low levels of self-control are more likely to engage in criminal/deviant behavior when criminogenic opportunity structures are available, under certain circumstances (
Akers and Sellers 2013). Accordingly, once self-control is developed during childhood through parenting practices and socialization, the degree to which individuals possess the restraint to avoid criminal acts typically remains in place. Once developed, self-control remains the same throughout the life course and is resistant to change (
Akers and Sellers 2013).
A number of studies have explored the relationship between self-control, religion, and criminal offending (
Welch et al. 2006;
McCullough and Willoughby 2009;
Kerley et al. 2011;
Laird et al. 2011;
Reisig et al. 2012). In addition, the role of morality as a predictor of criminal offending has been examined (
Antonaccio and Tittle 2008). Collectively, available literature reinforces Gottfredson and Hirschi’s contention that self-control influences criminal propensity amongst individuals (
Welch et al. 2006;
Antonaccio and Tittle 2008;
Kerley et al. 2011;
Reisig et al. 2012). For instance, in their study,
Welch et al. (
2006) examined if the relationship between personal Christian religiosity (i.e., literalism or religious interpretation, salience or meaningfulness of beliefs, frequency of prayer, and attendance at religious services) and future misconduct (i.e., likelihood that respondents would engage in illegal gambling, petty theft, DUI, assault, and tax evasion in the foreseeable future) was spurious or associated with the variable self-control. More broadly, the focal point of their study centered on Gottfredson and Hirschi’s assertion that the associations between “social variables and force, fraud or analogous imprudent acts” emerge in part by the “antecedent effect of self-control” (
Welch et al. 2006, p. 1606). In this sense, the relationship between social variables such as religion and criminal offending is spurious in nature given the influence of self-control in aiding individuals to adhere to underlining expectations of one’s moral community and conform and safeguard social bonds (
Welch et al. 2006). Further, through self-control, individuals internalize and operationalize fundamental principles of their religion that further dictate and guide individual behavior towards conformity (
Welch et al. 2006).
Results from
Welch et al. (
2006) revealed that, in opposition to Gottfredson and Hirschi, the association between religiosity, a social variable, and future misconduct was not “spuriously attributable to prior levels of self-control” (
Welch et al. 2006, p. 1615). Instead, both religiosity and self-control were found to “operate independently of one another”, net of socioeconomic indicators (i.e., gender, race, age, educational level, family intactness during childhood, and types of place of childhood residence) (
Welch et al. 2006, p. 1615). This suggest that high levels of religiosity and self-control decrease the likelihood of future misconduct (
Welch et al. 2006).
In another study,
Kerley et al. (
2011) examined the degree to which the religion-crime relationship was mediated by levels of self-control. Using data collected from a stratified convenience sample of 208 male parolees, findings from
Kerley et al. (
2011) coalesced with Gottfredson and Hirschi’s contention that the relationship between religion and criminal offending is spurious, noting that levels of self-control serve as a compounding factor. Specifically,
Kerley et al. (
2011) found that self-control significantly predicted behavioral measures of religiosity (i.e., praying privately, watching religious television broadcasts, and attending religious classes or groups), as well, when regressed on prison deviance alongside behavioral measures of religiosity, criminal history, and inmate demographics, significantly and negatively explained inmates’ involvement in prison deviance. These findings suggest that the relationship between religiosity and prison deviance is partially moderated by levels of self-control. Furthermore, findings indicate that levels of self-control and attendance at religious classes or groups operate independently of each other to predict prison deviance, buttressing
Welch et al.’s (
2006) findings.
Nevertheless, despite the deterrent effects of self-control relative to criminal offending, available literature deviates from Gottfredson and Hirschi’s argument supporting self-control as the primary predictor of criminal offending (
Antonaccio and Tittle 2008). In their study,
Antonaccio and Tittle (
2008) sought to isolate the effects of morality (i.e., dynamic moral beliefs and actions that are specific and contextual) and self-control relative to criminal propensity in order to gauge their relative predictive strength. For their study, religion, conceptualized as childhood religiosity (i.e., the degree to which respondents were religious as a child), served as a control variable. Based on findings from
Antonaccio and Tittle (
2008), although self-control has a statistically significant negative effect on criminal propensity, net of the control variables, morality possesses an even stronger association with criminal propensity, net of the control variables, that is also statistically significant. In another study, Reisig and colleagues (
Reisig et al. 2012) examined the degree to which the religion-crime association was spurious “after individual variations of self-control are controlled for in different multivariate contexts” (p. 1177). Conceptualized along three important domains of religiosity, religion was conceived as the frequency to which individuals engage in prayer and attend religious services (religious activity), the degree to which one’s religious beliefs are integrated in their daily lives (religious devotion), and individual endorsement of life after death (religious belief) (
Reisig et al. 2012).
Reisig et al. (
2012) found that, net of the control variables, all religious domains significantly reduced the likelihood that individuals would engage in ascetic offending (i.e., individual gratification prohibited by most religious traditions and criminal law, yet are not completely shunned by society at large), independent of self-control. In the case of secular offending (i.e., law-breaking behaviors with strong norms against conduct expressed by religious and nonreligious individuals), all religious domains were insignificant, net of all control variables, with self-control emerging as the key significant predictor (
Reisig et al. 2012). Here, depending upon the nature of the offense, religion has a potential inhibiting effect that is independent of self-control.
The extant literature indicates that while self-control is a strong predictor of whether or not individuals will engage in criminal activity, religion also seems to play a significant independent role as well. The integration of beliefs, values, attitudes and ideologies associated with a particular faith can curtail involvement in criminal activity if these traits are an integral part of the individual’s psyche. Furthermore, extant literature suggests that additional variables, such as morality, have a greater potential for predicting the inverse religion-crime relationship than self-control; self-control is not the only variable capable of predicting the religion-crime nexus.
3.2. Religion and Social Control
While religion may act as a form of self-control, it can also be argued that certain aspects of religion work as a form of social control as well. In examining religion as a form of social control, social scientists typically point to the ways in which being a member of a religious community may exert influence over decisions and behaviors people engage in. In this sense, control over behaviors has more to do with meeting the expectations of the group and maintaining a connection to the group. It is understood that failing to adhere to certain behaviors becomes a violation of the religious group’s expectations. Thus, the reason individuals may shy away from behaviors that are considered deviant, is to avoid being ostracized by the group. In addition, it has also been argued that participation in religious activities serves as a form of social control as well. This is particularly true when these activities are engaged in a group setting. Group participation in religious activities can be a powerful influence over behavior.
One of the aspects of religion that Durkheim examined in his work is the notion of the collective conscience. For Durkheim, collective conscience is the “totality of beliefs and sentiments common to average citizens of the same society” (
Ritzer and Stepnisky 2017, p. 83). Because Durkheim saw this collective conscience as a means to determine certain social facts, religion can be considered as a social fact that may or may not be a guiding factor in the lives of individuals. Therefore, religion, as a part of society’s collective conscience, can exercise certain forms of social control over the society’s members. Durkheim (
Ritzer and Stepnisky 2017) suggested that the social facts of integration, the “strength of the attachment that we have to society” and regulation, which refers to the “degree of external constraint on people” (p. 93) would determine the strength of social control over a society. Thus, integration and regulation could be applied to religion and the amount of social control that can be experienced.
Because individuals are so closely connected to religion, we would assume that religion can be a powerful influence in the lives of society’s members. With the influence of religion, we could also assume that social control exerted by religiosity would prevent individuals from taking part in criminal activity.
In his study of suicide, Durkheim addressed the diverse types of religions, with those possessing more exclusivity having lower suicide rates (
Durkheim 1951) due to the increased social control over the lives of individuals. As well, in
Durkheim’s (
1951) study of suicide, he found that religions that provided individuals with greater freedom, less social control, and “fewer common beliefs and practices” (p. 159) would have higher rates of suicide. These factors show the amount of social control some religions have over suicide, more specifically, but generalized criminal activity, more broadly. Recent studies have also examined the relationship between religion, social control, and criminal activity.
For example,
Johnson et al. (
2001) and
Desmond et al. (
2008) examined samplings of youth from the National Youth Survey (NYS) to determine if religiosity and moral beliefs had any influence on delinquency. Both studies (
Johnson et al. 2001;
Desmond et al. 2008) hypothesized that social bonds held the key to youth involvement, or the lack of, in delinquent behavior. Both studies (
Johnson et al. 2001;
Desmond et al. 2008) examined the theoretical elements of social bond theory (i.e., attachment, commitment, involvement, and belief) and the strength of these bonds as the guide for delinquent behavior. Findings from
Desmond et al. (
2008) revealed a strong connection between the socialization of youth and “strong moral beliefs” (p. 66). Specifically,
Desmond et al. (
2008) found that youth who were socialized to believe that criminal behavior, in any form, ranging from stealing to drug use, was “morally wrong” (p. 66) abstained from that type of behavior. According to
Johnson et al. (
2001), due to youth commitment to his/her religious beliefs, that “religiosity reduces delinquency partly because religious involvement increases his or disapproval for delinquent acts” (p. 38) as well as reduces the connection of youth to delinquent peer contacts.
Similar findings were presented by
Knudten and Knudten (
1971) in their review of previously conducted research on religion and crime. Knudten and Knudten found that the perceived “criminal deviance is a by-product of the unstated assumptions of our society” (
Knudten and Knudten 1971, p. 131), and those assumptions were the integral part of the perceptions of crime in our modern society, stemming from the Puritan beliefs of the country’s early history. As a result,
Knudten and Knudten (
1971) questioned the amount of social control fostered by religion and how those religious insights would affect public laws. Their research did find that, in many instances, fanatical religious groups would indeed abstain from any type of criminal activity, which would also prove the sustainability of social control religion holds over a society.
In another study,
Curry (
1996) used data collected from a 1993 study conducted by the Department of Sociology, University of Oklahoma, to measure if religion influenced a “perceived wrongfulness of crimes” (p. 453). According to
Curry (
1996), religion and punishment were closely connected, but his study investigated if there was an additional relationship between religion and the perceptions of crime in religious communities. The results of the
Curry (
1996) research did reveal “that conservative Protestants tend to view all criminal behavior as very wrong” (p. 462). Also,
Curry (
1996) found very little difference in these beliefs between minor crimes such as trespass to more serious crimes such as murder. Although research by
Curry (
1996) was atheoretical, the social control in a society, such as that of a conservative religious community, would have the most profound influence over that society. Hence, this conclusion is aligned with Durkheim’s collective conscience of a group as well as Durkheim’s notion of integration of beliefs within a group. Each of these factors influences whether individuals commit or do not commit crimes.
Later in 1999, Sherkat and Ellison examined the sociology of religion by reviewing academic journals that were considered prime in their fields during the time of the literature review. While conducting this examination,
Sherkat and Ellison (
1999) consistently found that socialization among religious families, along with “(b) gender, (c) social status, and (d) life course events and aging” would have a direct influence “through the socialization of beliefs and commitments” (p. 367). This review of the literature also provided a substantial support for the element of commitment in social bond theory to again steer individuals away from criminality. As well, similar to previous research by
Knudten and Knudten (
1971) and
Curry (
1996), social control exhibited by the more conservative religious groups created an atmosphere whereby criminality, in any form, would be considered wrong and group members would be less likely to participate in crime in any fashion. At the conclusion of the
Sherkat and Ellison (
1999) review, the reviewers noted that more research was needed to determine if “religious value is collectively produced versus when it is a private good with intrinsic value” (p. 386), further supporting Durkheim and the power of collective conscience and religiosity (
Durkheim 1951;
Ritzer and Stepnisky 2017).
Contemporary scholarship examining the relationship between religion, social control, and criminal activity further illuminate the inhibitory role of religion on criminal/deviant propensities relative to social control variables. As an agent of social control, religion, as argued by
Cochran et al. (
1994), fosters conformity and reduces the likelihood for crime/deviance by “encouraging the internalization of moral values and acceptance of social norms” (p. 93). However, the degree to which religion alone decreases individual propensities for crime/deviance remains questionable. Hence,
Cochran et al. (
1994) assessed whether the religion-crime relationship was spurious by examining the effect of religious variables (i.e., religious participation, religious salience, and religious affiliation) on various indicators of delinquency (i.e., interpersonal delinquency, property-theft, property-damage, skipping classes, and use of drugs and alcohol), net of arousal (i.e., self-reported thrill-seeking, impulsivity, and physicality), and social control (i.e., internalized controls, parental controls, and institutional controls) variables. Findings from
Cochran et al. (
1994), though supportive of the inverse relationship between religion and delinquency, revealed that the religion-crime nexus was mediated by social control variables. Similarly,
Brauer et al. (
2013), in examining the validity of the inverse religion-crime relationship and assessing the effects of self-control, informal social control, morality, negative emotion, and social support on criminal probability, found that the inverse relationship between religion and crime was likely explained by theoretical mechanisms of self-control, informal social control, and morality.
Reinforcing
Stark’s (
1996) moral communities thesis, which posits that religion be sociologically studied at the macro-level as a group process and not that of individuals,
Regnerus (
2003) investigated the risk of delinquency amongst adolescents residing in religiously homogenous schools and counties that were characterized as being “above average social disorganization” (p. 529). In this study, religiosity was measured as adolescent’s self-identified level of church attendance and their identity as “born again” Christians, while individual-level measures of social control included, school attachment, family satisfaction, and autonomy from parents (
Regnerus 2003, p. 530). At the country-level, religiosity was measured in terms of church membership and adherence, as well indicators of social disorganization (i.e., family disruption, female headed-households, residential mobility, proportion children five years of age and older, proportion of Black population) and school-level factors (i.e., proportion of students from broken homes, proportion of students engaged in sexual intercourse, proportion of Black students, and level of intergenerational closure) were included (
Regnerus 2003). Though not explicitly stated, findings from
Regnerus (
2003) suggest that the religion-crime relationship is spurious given that conservative Protestant homogeneity in schools and counties not only served to shield adolescents from engaging in delinquency, but further mediated the inverse relationship between religion conceived as self-identification as “born again” Christians and delinquency (i.e., theft).
Overall, available literature underscores the protective role of religion as an agent of social control, as well reveals the interactive nature between religious practices and informal agents of social control in reducing criminal/deviant propensities. Nevertheless, though these studies contribute to the growth in epistemologies within sociological and criminological empirical enterprises, these studies are not immune to having empirical drawbacks.