About The ARDA | Tools | FAQs | Contact Us

In the social sciences generally, as well as in the social science of religion, the term theory is actually used in a multitude of applications. In a sense, every specific theory embodies a somewhat different idea of what theory means, so it is not surprising that this word tends to confuse people. For example, fully 93 articles in the Stanford Encyclopedia of Philosophy have “theory” in their titles, yet they approach it from almost as many different directions.

Citing the work of Rodney Stark and William Bainbridge, we offer the following general definition of a theory:

theory is a set of statements, or hypotheses, about relationships among a set of abstract concepts. These statements say how and why the concepts are interrelated. Furthermore, these statements must give rise to implications that potentially are falsifiable empirically.


Rodney Stark and William Sims Bainbridge, A Theory of Religion (New York: Toronto/Lang, 1987), p. 13.
The following Theories are available.

Human beings may have always believed that outsiders to the community are less trustworthy than community members, but only in the twentieth century was this observation formalized — in Control Theory.  The pivotal statement of the theory was offered in the book Causes of Delinquency by Travis Hirschi (1969), who used questionnaire data augmented by police records to illustrate and test the proposition that high school boys are less likely to commit delinquent acts if they have strong social bonds.  But he went beyond that simple idea to identify four aspects of social connectedness:

  1. Attachment: Strong connections to other people, especially parents, that make a person sensitive to their wishes.
  2. Commitment: Investment of time and energy in a conventional line of activity that would be endangered by delinquency.
  3. Involvement: Being so busy in conventional activities, such as adult-organized recreation, that there is no time free to perform delinquent acts.
  4. Belief: The conviction that the norms of society are good and should be obeyed.

Religion could play a significant role in any of these, and of course involvement in church-based activities is especially obvious.  However, when Hirschi tested these four aspects of the theory, he found evidence for only the first three.  Legend has it that at this point, Rodney Stark asked Hirschi why he had not highlighted in his book manuscript the role religion played in preventing delinquency.  Hirschi replied that his data did not reveal any such normative power of religion for adolescents, and that led to their influential journal article “Hellfire and Delinquency” (Hirschi and Stark 1969).  Years later, Stark (et al. 1982) returned to this issue to explore possible geographic variations in the power of religion to deter delinquency.  The main finding may be called the Stark Effect: The power of religion to deter delinquency is significant when a substantial fraction of the population is religious, but absent when only a minority belong to religious organizations.  This reflects a very interesting general issue for social theory, in that the connections between variables within the minds of individuals may be conditioned by the organization of the surrounding society.  That is, in the part of California where Hirschi’s data were collected, churches are weak, so even boys who are believers may confine their faith within the walls of the church and ignore religious principles when they are in secular environments.

The Stark Effect also reminds us of the prior history of Control Theory.  It was a condensation of the dominant perspective on deviant behavior of the Chicago School of Sociology (Papachristos 2012), in the third of a century before Hirschi wrote, that identified its cause as social disorganization.  Over the early decades of the twentieth century, the sociology department at the University of Chicago dominated sociology, with an intense focus on urban studies, often doing research on differences across the neighborhoods of Chicago itself.  A famous ethnography by Nels Anderson (1923) described the chaos in an area of Chicago he called Hobohemia, a “Bohemian” district populated by migrant workers and unemployed hobos.  Rigorous statistical studies looked at variations in rates of crime and delinquency (Shaw and McKay 1927).  A study that did not focus on religion, but with findings that have theoretical implications for religion, was The Gang by Frederick Thrasher (1927).   When the community at large becomes disorganized, small-scale social organization will arise spontaneously.  While that principle was central to Thrasher’s theory of gangs, it could be applied as easily to the emergence of social groups that are not at all criminal, including religious sects and cults.  If traditionally dominant religious organizations lose their coherence, the result may be an eruption of reform and revival movements within these organizations, and novel religious movements entirely outside them.

The classic example of control theory applied to religion is Emile Durkheim’s (1897) study of suicide.  Beginning decades before Durkheim wrote, social scientists had begun using geographic rates of suicide to develop and test theories, and his own contribution consisted primarily of providing a theoretical framework.  One of his key concepts was egoism, not to be confused with egotism but similar, referring to an unhealthy individualism that left a person vulnerable.  He argued that the fragmentation of denominations and sects that marked Protestantism encouraged individualism, and thus increased the rate of egoistic suicides, compared with the more unified Catholicism.  Over the following century, sociological critics pointed out that Durkheim’s own data may not have supported his claim, for example because the high rates for Protestants seemed limited to German areas and thus may have reflected a different social cause.  But setting aside the debates over historical differences between Protestantism and Catholicism, more recent studies do seem to show that the coherence of religious communities is a factor deterring suicide (Pescosolido and Georgianna 1989).

Given how many variants of Control Theory exist, and Hirschi’s attempt to identify four subtheories within it, a survey of the social science of religion would identify many other models that could be given this label.  For example, Rosabeth Moss Kanter’s (1972) theory of commitment mechanisms is worth considering.  She developed it to explain the relative success of some nineteenth-century America utopian communities, which from her perspective only happened to be religious, although religious hope and faith may indeed have been required for these mechanisms to function.  Kanter identified six commitment mechanisms, which may be classified in three pairs.  The first one in each pair requires giving something up, while the second involves receiving something in return:


  1. Sacrifice
  2. Investment


  1. Renunciation
  2. Communion


  1. Mortification
  2. Transcendence

In many of the communes, members were required to give up ordinary material rewards, yet they were compensated by reasonably comfortable living conditions.  This could be considered in plain economic terms: they had sacrificed all their money to the commune, and would profit from this investment only if they remained loyal members.  They had renounced many social relationships with outsiders, but enjoyed an intense communion with fellow members.  They had accepted harsh criticism of the secular identities they had abandoned, but gained new and positive identities within the ideal community.  This last point did not necessary require mortification of the flesh, but of their former self-images, to achieve transcendence of the human condition, which only religion seems capable of offering.  This model of commitment to religious communities may be applied to ordinary churches, but with milder versions of each of the six mechanisms.  The first in each pair dissolves the control that non-religious society may have inflicted, while the second imposes control by the religious organization.  The vast literature on deviance, conformity and religion may include many plausible multi-component control theories, of which those by Hirschi and Kanter are prominent examples.


Anderson, Nels.  1923. The Hobo.  Chicago: University of Chicago Press.

Durkheim, Emile.  1897.  Suicide.  New York: Free Press (1951).

Hirschi, Travis.  1969.  Causes of Delinquency.  Berkeley, CA: University of California Press.

Hirschi, Travis, and Rodney Stark. 1969.  “Hellfire and Delinquency.” Social Problems 17:202-213.

Kanter, Rosabeth Moss.  1972.  Commitment and Community: Communes and Utopias in Sociological Perspective.  Cambridge, MA: Harvard University Press.

Papachristos, Andrew V.  “The Chicago School of Sociology.”  Pp. 472-479 in Leadership in Science and Technology, edited by William Sims Bainbridge.  Thousand Oaks, CA: SAGE.

Pescosolido, Bernice A., and Sharon Georgianna. 1989. “Durkheim, Suicide, and Religion: Toward a Network Theory of Suicide.” American Sociological Review 54:33-48

Shaw, Clifford, and Henry D. McKay. 1927. Delinquency Areas. Chicago: University of Chicago Press.

Thrasher, Frederic M. 1927.  The Gang.  Chicago: University of Chicago Press.

Stark, Rodney, Lori Kent, and Daniel P. Doyle.  1982.  “Religion and Delinquency: The Ecology of a ‘Lost’ Relationship.” Journal of Research in Crime and Delinquency 18:4-24.



Proposed by criminologist Edwin Sutherland, Differential Association Theory has been mainly applied in social science studies of deviant behavior, so it is reasonable to extend it to analysis of cults and sects, but Sutherland really proposed it as a general theory of social influence that could be applied very broadly.  The standard version was published in the fourth edition of his textbook, Principles of Criminology, and postulated that criminal behavior was chiefly learned within intimate personal groups (Sutherland 1947: 6-7).  An important part of that learning could be conceptualized as values, whether to view legal codes favorably or unfavorably, but also could include practical skills such as techniques for committing a crime.  Sutherland explicitly stated that his theory described the ways people learned all forms of social behavior, and that criminal behavior served the same needs and values as other kinds of behavior.

A central but frequently misunderstood principle of the theory was stated thus: “A person becomes delinquent because of an excess of definitions favorable to violation of law over definitions unfavorable to violation of law.”  Many scholars and students read the theory as if it were simply a model of social influence: If most of the people you associate with are criminals, you will be one, too.  But that is not at all what the theory is about, as the eleventh edition of Principles of Criminology clearly states (Sutherland et al. 1992: 92).  It is a theory of communication, what today would be included in information science, and was written within that sociological school of thought called symbolic interactionism.  The word definition repeatedly used by Sutherland connects to the work of W. I. Thomas (1923:42), who stressed the importance of the definition of the situation in determining human behavior.

For many years, beginning with Travis Hirschi’s 1969 book Causes of Delinquency, criminologists vigorously debated the relative merits of Differential Association Theory against Control Theory which Hirschi favored.  In Control Theory, people become deviant largely because their social attachments to other people are weak, so this battle of the explanations tended to support the revisionist version of Differential Association Theory that focused on differential social ties with criminal subcultures versus with conventional society.  Hirschi seemed to demonstrate that his own data, from a survey of California adolescents, entirely supported Control Theory, but the debate was never really settled, and at least one prominent criminologist found solid evidence in favor of Sutherland’s theory in Hirschi’s own dataset (Matsueda 1982).  Of course, there is no reason why both theories could not be correct, combining to explain deviance better than either could alone, especially when we recognize that Sutherland’s theory is not just about social bonds between people, but about the differential transmission of information and values.  Notably, the theory can give religion a prominent role to play in promulgating norms, the obvious example being the influence of powerfully expressed church sermons in convincing churchgoers to obey the laws of the society.

Thus, Differential Association Theory is about mental associations between concepts, and how people learn definitions of situation in complex patterns of communication with other people.  Therefore it can be adapted to model phenomena other than crime, especially when a complete cultural consensus is lacking, and people may receive communications that define things in different ways.  Sutherland expressed his theory in terms of nine propositions, which may easily be translated to describe the dynamics of religious belief and adherence (Bainbridge 2006:47-48):

  1. Religious behavior is learned.
  2. In interaction with other persons in a process of communication.
  3. Principally within intimate personal groups.
  4. The learning includes: (a) religious practices, and (b) religious beliefs, attitudes, and values.
  5. The religious beliefs, attitudes and values are learned from definitions favorable to acceptance of the particular religion or unfavorable.
  6. A person converts to a religion because of an excess of definitions favorable to the religion over definitions unfavorable to the religion.
  7. Differential associations may vary in frequency, duration, priority and intensity.
  8. The process of religious learning involves all the mechanisms that are involved in any other learning.
  9. While religious behavior is an expression of general needs and values, it is not explained by those general needs and values, since nonreligious behavior is an expression of the same needs and values.

The seventh point deserves special scrutiny.  Suppose a person is receiving mixed messages about a religious belief, both pro and con.  If someone says the belief is true, that associates the belief with truth.  If someone else says the belief is false, that associates the belief with falsehood.  The differential association is the balance between these two kinds of message.  If the favorable association is communicated with greater frequency, then it outweighs the less frequent negative association.  The duration variable suggests that a long message, like a church sermon lasting twenty minutes, will outweigh a quick comment by someone who rejects the church.  Priority concerns earlier versus later messages, for example the power of religious instruction in childhood.  Intensity, in the context of symbolic interaction, may simply be how much emotion is invested in the message.

Aspects of this theory are quite compatible with other communication-related studies that date from about the time Sutherland was working, that have stood the test of time, and that can be applied to religious topics.  For example, in the 1950s many researchers were studying the influence of the mass media, which seemed to be able to influence citizens in an impersonal manner.  Or, the mass media were impersonating personal influence, as people might come to feel they had a personal relationship with television newscasters and actors.  The most relevant example is Jack Webb, who created the highly popular 1951-1959 TV crime drama, Dragnet, in which he himself played the role of a policeman named Joe Friday, who always triumphed over the bad guys in 276 episodes of this half-hour program.  Among the most significant sociological works of that decade was Personal Influence by Elihu Katz and Paul Lazarsfeld (1955), which offered a two-step model of diffusion of information and values.  Whether from the mass media, or some more obscure source, a few open-minded or creative members of the local community would become enthusiastic about something, then promote it through the immediate community, serving as opinion leaders.  Both the moralism dramatized in Dragnet and the enthusiasm of a local opinion leader increase the intensity of communication of norms, thus being quite compatible with Differential Association Theory.

Many social scientists have offered critiques and extensions of Sutherland’s theory.  Robert Burgess and Ronald Akers (1966) revised the theory considerably, removing it from the symbolic interactionist tradition and situating their version in Behaviorism.  This was a huge modification, implying that people accept a religious belief if they receive tangible rewards for doing so.  In their reformulation, differential association becomes the cost-benefit ratio, thus rather more relevant for understanding why people might steal property than why they pray for guidance, but not totally incompatible with use of the theory to explain at least some kinds of religious behavior.

In recent years, the Internet has revolutionized the mass media, fragmenting or diversifying the flood of information and norms, and giving people technologies that allow them to communicate 24/7, and that create online communities that may sometimes transmit intense definitions.  Thus, we seem to have entered a period in which Differential Association Theory is not only relevant but needs to be updated and expanded through a variety of innovative empirical research projects.


Bainbridge, William Sims.  2006. God from the Machine: Artificial Intelligence Models of Religious Cognition. Walnut Grove, CA: AltaMira Press.

Burgess, Robert L., and  Ronald L. Akers.  1966. “A Differential Association-Reinforcement Theory of Criminal Behavior,” Social Problems 14(2): 128-147.

Hirschi, Travis.  1969.  Causes of Delinquency. Berkeley, CA: University of California Press.

Katz, Elihu, and Paul F. Lazarsfeld.   1955.  Personal Influence: The Part Played by People in the Flow of Mass Communications.  Glencoe, ILL:  Free Press.

Matsueda, Ross L.  1982.  “Testing Control Theory and Differential Association: A Causal Modeling Approach,” American Sociological Review 47:489-504.

Sutherland, Edwin H.  1947.  Principles of Criminology.  Chicago: Lippincott.

Sutherland, Edwin H., Donald R. Cressey, and David F. Luckenbill.  1992.  Principles of Criminology.  Lanham, MD: General Hall.

Thomas, William I.  1923.  The Unadjusted Girl.  Boston: Little, Brown.

This kind of theory postulates that the human mind naturally seeks simple models of reality, and that humans will tend to avoid extreme cognitive effort. This perspective is similar to, but distinguishable from, cognitive consistency theories.

The human mind can hold a vast number and variety of memories, from episodic recollections of specific past events and separate but connected glossaries of words in multiple languages, to manual skills like typing on a keyboard in which the letters of the alphabet are arranged in an illogical sequence.  Yet when guiding behavior, relatively few pieces of information seem to dominate, thereby facilitating quick and decisive action.  In 1956, psychologist George A. Miller observed that humans can hold about seven items in short-term memory, also called working memory, and can assign sensory inputs such as musical tones to a maximum of about seven categories.

Vast amounts of information may pass through short-term memory during the course of a life, but only a small fraction of these data persists beyond a few minutes, for example building up autobiographical memory over years (Conway 2001). Miller was aware that the human brain fed information for long-term memory into short-term memory in what cognitive science came to call chunks, and chunking was one of the key methodologies developed for computer-based artificial intelligence over the subsequent decades (Newell 1990).  In reflecting upon one’s religion, at one moment a particular moving church service may be recalled, but at another moment the particular words of a verse in the Bible or the visual image of a crucifix may dominate consciousness.

In 1954, Gordon Allport offered a cognitive effort theory of prejudice, based on this axiom of cognitive efficiency.  In a study that frequently used examples from religion, he said that humans tended to conceptualize the world in terms of simple categories that require little effort to apply, rather than, for example, multiple dimensions of continuous variation.  Humans tend to classify other humans into distinct groups and assume that every member of a group shares the same primary characteristics.  This applies to stereotypes of members of religious groups other than our own.  While only empirical research can determine the extent to which this theory constrains human cognitions about other faiths, it has proven relatively easy to program artificial intelligence systems that can solve problems relatively well, while consolidating into suboptimal models of reality comparable to human prejudice (Bainbridge 1995).

Pascal Boyer (2001) and Justin Barrett (2004) argued that supernatural ideas must be minimally counterintuitive if they are going to be popular.  That is, they must contradict ordinary assumptions about life in only one or two important respects.  God, for example, is conceptualized as a person, having thoughts and desires just like humans, but also possessing special powers.  A minimally counterintuitive idea is by definition an example of cognitive simplicity.

If human short-term memory evolved to maximize cognitive efficiency and thus provide quick and decisive control over real-time human action, then it may not only be limited but also structured in a way that is oriented toward action.  H. Porter Abbott (2003) suggests that human thought organizes things in terms of narratives — stories in which protagonists face obstacles and take actions in pursuit of goals — and thus the scientific theory of evolution by natural selection from random variation cannot compete for popular acceptance with religious stories because it is unnarratable.  Large fractions of the holy scriptures of the world are stories, and deities often are framed as personified forces of nature.  It also is possible that belief in an afterlife reflects the fact that we develop mental models of the feelings, beliefs and intentions of the individual people closest to us, and the models persist after the deaths of the people (Bering 2006).

Cognitive efficiency theories can be applied to most domains of human endeavor, including the social science of religion itself.  The obvious example is Max Weber’s (1949) methodology of conceptualizing human thought, action and social organization in terms of ideal types, when he introduced the church-sect duality into the sociology of religion.  Harvey Whitehouse (2004) has proposed a cognitively simple cognitive theory of the difference between a church and a sect.  Mainstream religion, relatively intellectual and involving a host of symbols, is rooted in semantic memory that learns through constant repetition of logically connected ideas.  Sectarian religion is more emotional and rooted in episodic memory that is activated by emotionally intense but rare religious experiences.  The two modes of religiosity, according to Whitehouse, spread socially through very different forms of activity in the brain that are functionally distinct and may be anatomically distinct as well.

On the basis of his fundamental postulate that humans adopt behavioral strategies that reduce cognitive effort, it is noteworthy that Gordon Allport (1960) developed a very influential distinction between two types of religious motivation, extrinsic and intrinsic.  Extrinsic religiousness is utilitarian, using religion to obtain rewards of value in ordinary life, even supernatural rewards directly from God.  Intrinsic religiousness is not greedy in this way, but reflects sincere adherence to the faith, even accepting the unselfish goals prescribed by it.  Allport considered intrinsic religiousness to be more mature, thus connecting this distinction to theories of human cognitive development.  Yet given how much he published about religion, Allport might himself be considered religious, and framed his theories of human cognition at least in part to harmonize with his religious beliefs.

It is possible for an ideal type theory of religion to be objective, rather than an expression of academic or theological prejudice, if it is based on logical alternatives.  For example, Andrew Greeley (1989) postulated the cognitive theory that Catholics tend to have analogical imagination, while Protestant have dialectical imagination.  The analogical way of thinking believes that God is present in the world, expressing Himself through every aspect of creation, and it stresses the community.  The dialectical imagination believes that God has withdrawn from the sinful world, and it stresses the individual.  Of course, the metaphor of withdrawal suggests movement away from us along a spatial dimension, not a leap from inside our world to outside.  Also, Greeley’s dichotomy seems not to apply well to other religious traditions, notably those that are polytheistic.

Jupiter used to visit our world frequently, and interacted directly with living humans, while Neptune and Pluto were as distant from us as those astronomical objects that bear their names.  Thus classical paganism could be described as both analogical and dialectical. It would at least appear superficially that polytheisms are cognitively more complex than monotheism, perhaps because they represent a stage in the unification of separate tribes of people who worshiped different gods, or because they represented different features of nature that required centuries of cultural development to bring together.  Stark and Bainbridge (1987: 55-87) considered at some length the proposition that the evolution of religion would be toward greater simplicity, in part to avoid the dangers of magic, which falsely appears to serve Allport’s extrinsic motivations, but also reflecting the gradual unification of many smaller social systems into a few large one.  However, they cautioned against the too-simple deduction that the end result would be pure monotheism, because, to put the point in Greeley’s terms, that could mean total withdrawal of God from the world, which would render Him simply irrelevant.

A very different critique of the cognitive simplicity theory was proposed by H. Porter Abbott (2013) and suggested by the title of his book, Real Mysteries: Narrative and the Unknowable.  While drawing upon twentieth-century existentialism, his argument restates a very traditional perspective fundamental to some Asian religions, notably Zen Buddhism, but incorporated worldwide in mysticism: Much of human experience cannot be put into words, and thus cannot be captured cognitively.  Miller focused on the chunked contents of short-term memory, ignoring the vastly more complex and even ineffable contents of the entire brain.  Allport’s extrinsic religion does so as well, and the lack of clarity to his concept of intrinsic religion may merely indicate it is stored outside short-term memory.  Thus, cognitive efficiency theory may explain much, but can only with difficulty explain some of religion’s fundamental qualities.

Attachment theory posits that religion can be explained by understanding the human need for attachment in general, and one’s relationship to her/his parents specifically.  Childhood experience is of particular importance to explaining adult religiosity from this perspective.  Attachment styles are generally divided into a tri-chotomy of:

  1. Secure,
  2. Avoidant, and
  3. Anxious/ambivalent.

Parental relationships are implicated in religiosity in one of two ways:

  1. Modeling, where one’s human attachment style serves as the pattern for attachment to religion and the divine or
  2. compensation, where religion serves as a substitute for human relationships.

Within the framework of the theory, the need for attachments is underpinned by biosocial impulses and desires that have developed over the period of human evolution.

Used in the psychology of religion, these approaches attempt to delineate the different motivations for the expression of religiosity.  Originally developed by Allport and Ross (1967), the effort was to distinguish internal motivation to be religious from the external or immediate benefits offered by participation in a religious community.  In other words, it seeks to distinguish between “religion that serves as its own end goal” and religion “used in the service of other goals or needs–i.e. as an instrumental value” (Gorsuch 1988: 210).

This research led to many future studies employing and refining the scale, most notably dividing the extrinsic dimension in two, as well as the development of metrics to assess different orientations such as Batson’s (1976) “quest” dimension.  Debates remain about whether some actions, such as attendance at worship service, are indicative of intrinsic or extrinsic religious motivation.

Originating in psychological studies of religion, these approaches highlight the ways in which individuals utilize religion to cope with difficult situations and make sense of events in their lives.  Research indicates that people typically have naturalistic and religious meaning systems operating simultaneously, and how these systems are combined to make sense of events depends on the characteristics of both the individual and the event in question.  Research on attributions is typically carried out in experimental settings where subjects are presented with scenarios or vignettes and asked how they would explain and understand the situations presented.

Similarly, research and theory indicates that religious coping is more likely to occur in situations perceived as uncontrollable.  As such, levels of religious coping vary not only by individual characteristics, but also by social location (status set).  Together these theories posit that religion helps satiate human need for meaning and explanations addressing existential concerns.

Interactive Ritual Chain Theory (IRC) is a perspective focusing on the interactions and the emotional input and feedback of individuals within those interactions.  Drawing on the work of Durkheim ([1912] 1995) and Erving Goffman (1959, 1967), the theory is formally posited by Randall Collins (1993, 2004; see his 2010 essay on micro-sociological approaches to religion).  IRC asserts that interactions produce or deplete the “emotional energy” of participants depending on key factors.  These factors include the physical co-presence of interactants, exclusivity of the group, a mutual focus and mood, and bodily synchronization.  Given its situational emphasis, IRC is difficult to measure quantitatively, but it is insightful for qualitative research, especially in areas such as worship rituals.  Importantly, IRC theory also situates religious actors in social space and outlines the linkages between ritual, affect, and belief.

This perspective posits that religion survives and can thrive in pluralistic, modern society by embedding itself in subcultures that offer satisfying, morally-orienting collective identities which provide adherents with meaning and a sense of belonging.

In a pluralistic society, religious groups which are better at transmitting and employing the cultural tools needed to create both clear distinction from, and significant engagement and tension with other relevant out-groups (short of becoming genuinely countercultural), will be relatively successful (Smith, 1998:118-119).  Although similar to the religious economies claim that groups at a medium level of sociocultural tension will attract the most adherents, this perspective focuses more on issues of identity and symbolic boundaries by drawing on cultural sociology.

Studies of conversion, religious schisms, and secularization utilize social network theory to understand the influence of community and networks on the religious life of individuals, groups and societies. In his classic study of suicide, Emile Durkheim used religion as an indicator of how well or poorly a society was socially integrated; other research indicates that religion actively builds social networks (Durkheim 1897, Stark and Bainbridge 1981, Bainbridge 1987, Bainbridge 2006).

Religion may produce strong social networks, but it also depends upon them. Thus religion may feature as either the independent or dependent variable in studies related to this theoretical perspective. There are a number of studies of social networks where religious vitality is the result of levels of social integration (Bainbridge 1990, Stark and Bainbridge 1980, Bainbridge 1989).  Also see “Conversion Theory.”

Secularization has a long history of theory relating to the idea that religion will become less powerful as a social institution with the progress of “modernity.”  For example, Berger defines the term as meaning: “The process by which sectors of society and culture are removed from the domination of religious institutions and symbols” (1967:107).  Secularization perspectives are varied, but in general there are three levels upon which secularization is theorized to occur (see Tschannen 1991).  These are: 1) Macro – social differentiation; 2) Meso – the decline of significance of religion in organizations; and 3) Individual – a reduction in levels of practice, belief, or affiliation at the individual level.  A hotly debated topic concerns the degree to which secularization is an inevitable process as societies “modernize,” or whether instances of secularization are “exceptions” (see Davie 2002; Martin 1991).  Another source of contention is whether all three levels of secularization are necessarily linked together or whether processes at one level may occur without those at another.

Secularization was for a long period the dominant perspective on religious change in the social sciences.  Yet there has never been a single theory of secularization.  A family of theories drew on classical sociological theory.   Durkheim posited that increasing social differentiation as a result of the expanding social division of labor would lead to the separation of the sacred and secular realms.   Secular institutions would become predominate, the “collective conscience” generated by religious participation would erode, and the functions performed by religion would be taken over by newly specialized institutions, such as the nation-state and the education system.  Weber proposed that the increasing dominance of instrumental rationality in economic and political institutions was leading to the “disenchantment” of everyday life, and eventually the eclipse of religious reason.  Marx saw religion as little more than an ideological system for the justification and perpetuation of class domination, arguing that as class consciousness and materialism advanced religion would disappear.

In general, contemporary proponents of various secularization theories observe that modernity tends to erode religion’s plausibility, intensity, and authority. Further, these theories tend to posit the contemporary retreat of sacred institutions, the privatization of faith, and the “progressive shrinkage and decline of religion” in public life (Casanova 1994).   Within this broad consensus, however, there are a variety of theoretical positions.

Peter Berger (1967) offers a micro-level version of the theory that focuses on the plausibility of religious concepts.   He argues that changes in religious consciousness are due not only to science and the Enlightenment but also to the expanding social and cultural pluralism that is a central feature of liberalizing societies.  This confrontation with pluralism was posited to damage the plausibility of religious dogma.  When religious adherents encounter credible others with rival and fully incompatible claims to ultimate truth, their own certitudes begin to suffer.  In liberal societies, multiple religious and secular groups jostle for influence on the basis of philosophical and ethical claims, undercutting each of their claims to predominance, and ultimately leading to the privatization of religion in civil society.

There is also a macro version of the theory. In the decades following the Second World War, most social scientists expected convergence among the Western industrial democracies and, with varying degrees of lag, a general move toward modernity by the “developing” societies. Modernization was expected to bring prosperity and opportunity, initiating a “culture shift” to post-traditional values and lifestyles (Inglehart 1990). As Ronald Inglehart and Wayne Baker (2000) explain, “Modernization theorists … have argued that the world is changing in ways that erode traditional values. Economic development almost inevitably brings the decline of religion, parochialism and cultural differences.”  In fact, cross-national research does suggest a generally negative association between development and religiosity, though with some caveats.  More recently, the modernization theory of secularization has been modified. Neo-modernization theory now rejects the linear implications of past formulations and seeks to link the micro and macro-levels in its explanation for religious change.  The key micro-level factor is now held to be existential insecurity; the greater that insecurity, the more likely that people will be religious.  But where economic, political and social conditions have improved such that personal security improves, religion loses its impact.  This does not mean that the world is growing secular, however, since most of the world’s population growth is occurring in the poorest, most unstable, and most traditional societies (Norris and Inglehart 2004).

There is no unified theory of secularization, and some of the mechanisms proposed by secularization theorists seem to remain obscure. In addition, it is not clear why indicators of “secularization” are high in some modern societies – for instance, in many Western European societies – but lower in others such as the United States, or why some countries that are far less developed than the Western industrial democracies are more irreligious.  Berger (2002), once a prominent proponent of the secularization thesis, now declares, “Our age is not an age of secularization. On the contrary, it is an age of exuberant religiosity, much of it in the form of passionate movements with global outreach.”  Social scientists proposing the religious economies model have generally been highly skeptical of secularization theory. Rodney Stark (1999) advised that the thesis be left to “rest in peace”; contrary evidence and theoretical shortcomings effectively consigning it to the dustbin of history.

Others note that secularization often appears to be an intentional political project, rather than a spontaneous socio-cultural development (Smith 2003; Froese 2009).  Many studies identify the central role played by church–state institutions in causing variation in secularization across societies. Political mobilization on the basis of religion is often triggered by the efforts of political elites to reduce the public role of religion (institutional secularization) or extend governmental authority to domains previously organized by religious organizations. State regulation or penetration into areas once dominated by religion often provokes conflict, especially where it threatens the influence of religious authorities. “Reasonable” government regulation of religious expression or activities in the interest of secularization is often viewed as an attack on religion.

Finally, there are the new cultural approaches to secularization theory that argue against linear secularization narratives but still contend that secularization is manifest in the pluralism of religious worldviews and highly individualized assemblages of religious and supernatural beliefs. As is becoming apparent in Europe and the United States, even where belief in the supernatural remains, denominational and confessional attachments appear to be weakening.  As a result, religious preferences are becoming more individualized, the status of orthodox religious authorities is diminishing, and growing proportions of people seek a spirituality divorced from conventional religion (Lambert 2004).  As Phil Gorski (2005) aptly puts it, “the weakening of traditional Christianity appears not as a decline of religion per se … but as a return to polysemism, since the new worldviews are not uniformly theistic.”  Thus it is not that religion inevitably declines but that religion as it has been theologically and institutionally understood in the West for the last several centuries is being irresistibly altered.

Religious economies perspectives are built on rational choice assumptions.  In particular these approaches assume that there are potential benefits to religious participation, both psychological and inter-personal, that people may seek.  With the analogy that overall levels of religion in an area are like a market, it assumes an innate demand for some type of explanation system and posits that differences in levels of participation can be at least partially explained by the supply of “religious goods” in an area.  Producers of these religious goods and individual consumers become central concepts, as well governmental regulation of religion.  Other important concepts include the level of tension between a religious group and the surrounding socio-cultural environment and the level of “strictness” required to be a member of a religious group.

“Individuals act rationally, weighing costs and benefits of potential actions, and choosing those actions that maximize their net benefits.” (Iannaccone, 1997:26)

“Within the limits of their information and understanding, restricted by available options, guided by their preferences and tastes, humans attempt to make rational choices.” (Stark and Finke, 2000:38) 

“A religious economy … encompasses all of the religious activity going on in any society.” (Stark and Finke, 2000:35)

This theory holds that religion is just as important a feature of modern society as it is of traditional society, but it takes different forms and possesses different characteristics. While compatible with functionalist theories, this theory does not depend upon them, because it concerns the historical transformation of religion, whereas religion’s functions may be constant. Among the most influential variants of this perspective is the pattern variable theory of Parsons and Shils.

According to them, five socio-cultural dimensions of variation together describe modernization and are applicable to religion as well as to all other major institutions of society. With the traditional end of the dimensions on the left, the pattern variables are:

  1. Affectivity – Affective Neutrality
  2. Self-orientation – Collectivity-orientation
  3. Particularism – Universalism
  4. Ascription – Achievement
  5. Diffuseness – Specificity

Modern religion, as defined by the right end of these dimensions, is emotionally cooler than traditional religion, which is oriented toward very large social collectivities such as all humanity rather than the self and which promulgates universalistic norms and hopes of salvation, placing responsibility for moral choice in the individual. Modern religion becomes differentiated from other institutions of the society. These continua are intended to be five dimensions, rather than aspects of one, implying that prior to the completion of modernization some societies and their religions might be mixed types.

A recent critique of this theory, which grants that it has been exceedingly influential, comes from Nils Gilman (2003).  Various forms of modernization theory are also intertwined with theories of secularization.  Although these links were at one time conceptualized in a teleological fashion (see Gorski and Altinordu 2008), more recent scholarship has suggested that the link between secularization and modernization is more complex (e.g. Almond et al. 2003 Casanova 1994; Warner 1993). 

According to this perspective, religion exists because it serves an integrating function for society as a whole. Durkheim suggested this when he argued that God represents the society, and in worshiping God, society really reveres itself. The elements of the culture that are essential to the society’s survival are labeled sacred, in this theory. Unlike theories of the rise and fall of civilizations, functionalists do not consider the survival of a religious culture to be problematic. While flavors of this theory are common in older writings on religion, among the best texts to consult are Durkheim 1915, Parsons 1937, and Parsons 1964.

Some readers of these works would draw the inference that religious individuals are more likely to follow society’s norms than non-religious individuals. However, pure functionalist theory does not rise or fall on whether this hypothesis is true, because it concerns the society as a whole, rather than individuals. In his 1964 journal article, Parsons suggests that the proof of functionalist theories of religion can be found in the fact that all societies have possessed religion – that it is an evolutionary universal necessary for the survival of society. He also considers religion to be a precondition for the development of many of the apparently non-religious features of modern society.

According to this theory, fertility and mortality rates change in a predictable manner, when a society evolves from a traditional to a modern form. Initially, death rates are high, because of the primitive technology and economic poverty of a traditional society, so birth rates were also high to sustain a stable population. At the beginning of modernization, technological and economic progress reduces the death rate, but the birth rate remains high through social inertia, so there is a population explosion. Eventually, the birth rate comes down as well, and the result is a stable population with low fertility and mortality rates. A classic statement comes from Kingsley Davis. 

By the mid-1980s, however, it was apparent that most advanced industrial or post-industrial societies had birth rates that were too low to offset even a greatly diminished death rate, and after the distorting effect of the changing age distribution had worked its way through, these societies seemed doom to population collapse. The immediate relevance to religion is two-fold, because low birth rates undercut some of the family-related variables that encourage religious participation and because one of the few factors that could sustain fertility at the replacement level is religion. However, as Nathan Keyfitz pointed out, only fundamentalist religions, like radical Islam, may have sufficient fertility, and thus society ironically may become more religious through modernization rather than less (Keyfitz 1987).

This perspective has been used to explain the relative success of more conservative religious movements, in spite of the high numbers of disaffiliates in post-industrial world.

Asian religions, and some classical western philosophers, believed that history consisted of an endless series of cycles: the Wheel of Life, eternal return, or eternal recurrence. This idea can also be found in nineteenth-century European philosophies that related to religion, notably the work of Friedrich Nietzsche, and it is not entirely implausible that processes analogous to the individual’s life cycle occur on larger, societal scales.

Probably the most impressive cyclical theory that gives religion a central role is the one proposed by Pitirim A. Sorokin. For Sorokin, the most influential elements of culture are those that concern the inner experience of people, their images, ideas, volitions, feelings, and emotions. The essence of a culture is defined by the view people have of the nature of reality, the goals they value, and the means they emphasize in reaching these goals.

In his theory, each great civilization emerges out of a period of chaos with a coherent set of spiritual beliefs that give it strength. Often it is born in the development of a new religious tradition. At this point, it is what Sorokin called an ideational culture. A successful ideational culture grows and develops. With success, however, comes complaisance. The society slowly loses its faith in spirituality, doubt sets in, and the culture begins to become sensate, a perspective on existence that is the opposite of ideational. A sensate culture believes that reality is whatever the sense organs perceive, and it does not believe in any supernatural world. Its aims are physical or sensual, and it seeks to achieve them through exploiting or changing the external world.

Depending upon circumstances, most people in a sensate society will exhibit one of three personality orientations. Active individuals use technology and empire-building to take charge of the material world. Passive individuals indulge themselves in pleasures of the flesh. And cynical individuals exploit the prevailing conditions for their own profit without any ideal to provide fundamental values. The entire cycle of which he wrote can take many centuries to complete, but Sorokin believed that western society was approaching a crisis point.

Ultimately, a sensate civilization is likely to crash, ushering in a new period of intense cultural chaos out of which a new ideational civilization may be born. Sorokin wrote, “Neither the decay of the Western society and culture, nor their death, is predicted by my thesis. What it does assert… is simply that one of the most important phases of their life history, the Sensate, is now ending and that we are turning toward its opposite through a period of transition. Such a period is always disquieting, grim, cruel, bloody, and painful” (vol. 3, p. 537).

Were Sorokin alive today, he probably would cite the rise of Islamic fundamentalism as confirmation of his theory, suggest that it would result in widespread conflict and religious revival in Islamic societies, even as European Christian culture continued to descend toward at least temporary collapse.

It deserves note that the notion of a cyclical periodicity has been also applied to Western religions, and to more narrow phenomena such as political movements.  For example, some researchers (Jelen 1991; Lienesch 1993) have suggested that religiously motivated political activity in the United States follows a cyclical pattern, based in part on the activation of negative affect toward out-groups (e.g. immigrants, homosexuals), and the subsequent mobilization of religious particularism.

Inspired by Lofland’s field research on recruitment to the Unification Church, in its earliest days sending evangelists from Korea to the United States, Conversion Theory offers a series of steps a person must go through in order to become a member of a new religious group:

  1. Experience enduring, acutely felt tensions
  2. Within a religious problem-solving perspective,
  3. Which leads him to define himself as a religious seeker;
  4. Encountering the group at a turning point in his life
  5. Wherein an affective bond is formed (or pre-exists) with one or more converts;
  6. Where extra-cult attachments are absent or neutralized;
  7. And where, if he is to become a deployable agent, he is exposed to intensive interaction (Lofland and Stark 1965).

Fifteen years after he had collaborated with Lofland, Stark worked with Bainbridge on a paper that implied that several of the steps of the model were unnecessary, and that frequency of social interaction – the last three steps – could be quite sufficient (Stark and Bainbridge 1980).  Although the process of conversion remains an important idea, the importance of social networks for conversion is the aspect of conversion theories that have had the most impact on further research.  More recently scholars have taken the focus on networks and incorporated it into religious switching, and even de-conversion.

“Cognitive science is the interdisciplinary study of mind and intelligence, embracing philosophy, psychology, artificial intelligence, neuroscience, linguistics, and anthropology.” Paul Thagard in the Stanford Encyclopedia of Philosophy, available online.

Recent cognitive theories have not yet been fully integrated into the social science of religion, but researchers have many opportunities to expand the scope of cognitive science (Bainbridge 2006).  Wuthnow (2007) provides an introductory overview of the issues, extant literature, and potential for research avenues with regard to cognitive theories and religion.

There are actually several distinguishable cognitive theories of religious phenomena, they are: 

  1. Attribution of Intentionality: Perhaps the most widely influential theory currently in the cognitive science approach to religion holds that faith in supernatural beings is a cognitive error that naturally springs from the way the human brain evolved (Boyer 2001, Barrett 2004, Abbott 2003).
  2. Cognitive Consistency: Theories in this sub-category argue that humans have a natural need to form consistent mental models of the world, and thus they will exert effort to resolve any contradiction between two beliefs, or between a belief and a behavior (Festinger et al. 1956, Festinger 1957).
  3. Cognitive Efficiency: This kind of theory postulates that the human mind naturally seeks simple models of reality, and that humans will tend to avoid extreme cognitive effort. This perspective is similar to, but distinguishable from, cognitive consistency theories (Miller 1956, Newell 1990, Boyer 2001, Allport 1954, Bainbridge 1995).
  4. Modes of Memory: Harvey Whitehouse has argued that different styles of religion are based in different parts of the brain, specifically in different memory structures (Whitehouse 2004).
  5. Pragmatic Epistemology: Whereas some theories consider religion to be the result of cognitive errors, this theory argues that religion serves the interests of the individual and thus is true for that very reason (James 1948).

Theories in this broad category assert that each major civilization, and perhaps smaller units as well in prehistoric times and remote regions, has a degree of cultural coherence, often marked by a distinctive religion. When two such civilizations come into contact, they compete, sometimes for several centuries, with resultant religious conflict. Also, it seems likely that every civilization eventually will exhaust its central cultural principles and collapse. Thus these theories tend to concern the rise and fall of civilizations (See Gibbon 1776).

Gibbon suggests that the establishment of Christianity as the state religion was an attempt by Constantine and some of his successors to strengthen the Roman Empire. More recent scholars have argued that Christianity did indeed have a characteristic that made it suitable as a stabilizer of the state and perhaps the civilization in which the state was embedded, namely particularism. By rejecting the truth of alternative religions, it asserted a principle of central authority that could be useful to stabilize the governance of a large society (See O’Donnell 1977).

A more general civilization theory was proposed by Oswald Spengler. Spengler asserted that every great civilization was based on a single idea, and when this idea became exhausted, the civilization would fall. For modern, Western civilization the idea was boundless space, and when the Age of Exploration came to an end, by the year 1900, the doom of the west approached. Ideas like Spengler’s continue to be popular among politically conservative intellectuals, some of whom consider Christianity to be the central idea of the West (See Burnham 1964, Buchanan 2002).

Among the modern attempts to develop data that could test or refine civilization theories, the World Values Survey stands out. Publications based on it tend to give a mixed picture, with some evidence that major cultural blocs in the world do indeed have somewhat different values, but perhaps not markedly different (See Inglehart and Baker 2000).  Samuel Huntington was well-known advocate of the idea that civilizations are based on competing and incompatible religious traditions, thereby limiting the extent possibilities concerning inter-civilization cooperation. 


Churches are religious bodies in a relatively low state of tension with their environments.

Sects are religious bodies in a relatively high state of tension with their environments.

The sect-church process concerns the fact that new religious bodies necessarily begin as sects or new religious movements and that, if they are successful in attracting a substantial following, they will, over time, almost inevitably be gradually transformed into churches. That is, successful religious movements will shift their emphasis toward this world and away from the next, moving from high tension with the surrounding socio-cultural environment toward increasingly lower levels of tension. As this occurs, a religious body will become increasingly less able to satisfy members who desire a high-tension version of faith. As discontent grows, these people will become dissatisfied that the group is abandoning its original positions and practices. At some point this growing conflict within the group will erupt in a split, and the faction desiring a return to higher tension will found a new sect. If this movement proves successful, over time it too will be transformed into a church, and once again a split will occur. The result is an endless cycle of sect formation, transformation, schism, and rebirth. The many workings of this cycle account for the countless varieties of each of the major faiths (Finke and Stark 1992:44-45).

The church/sect cycle is proposed as a general theory of change in religious organizations over time.  It is rooted in the work of Max Weber and Ernst Troeltsch, but has recently been taken up by rational choice theorists.

QuickSearch The Knowledge-Base

To search the knowledge-base, enter a term below:

Select a Theory below to learn more:

Select a Concept below to learn more: