CESNUR - center for studies on new religions
vilnius

RELIGION AND DEMOCRACY: AN EXCHANGE OF EXPERIENCES BETWEEN EAST AND WEST

THE CESNUR 2003 INTERNATIONAL CONFERENCE
organized by CESNUR, Center for Religious Studies and Research at Vilnius University, and New Religions Research and Information Center
Vilnius, Lithuania, April 9-12 2003  

A typology of new religious movements and its empirical indicators

by Tadeusz Doktòr, Warsaw University
A paper presented at the CESNUR 2003 Conference, Vilnius, Lithuania. Preliminary version. Do not reproduce or quote without the consent of the author.

Abstract

A two-dimensional typology of religious organizations proposed by Stark and Bainbridge (1987), related to the level of tension and innovation in relation to their religious environment was extended to include a fourth element representing loosely organized cultic milieu (with high level of innovation and low level of tension) and intrachurch religious movements. The dimension of tension was operationalized as an antagonism reflected in the difference between the general perception of Catholics and members of the own religious group. The dimension of innovation was operationalized as the measure of the degree of openness to other religious traditions. On the basis of research conducted among 643 members of different religious movements, mean values of these indicators were computed for each of the 24 movements. The localization of majority of these groups in two-dimensional space corresponds to the theoretical expectations based on this typology and fit to the predictions of Stark and Bainbridge theory of religion.

 

In their theory of religion Rodney Stark and William S. Bainbridge (1987: 17) have proposed a two-dimensional typology of religious organizations. According to these authors ‘a traditional idea like sect is best understood as the conjunction of two basic concepts from our theory – “conventional religious tradition” and “tension with the sociocultural environment”’. Each of these two concepts constitutes an axis of variation, which could be operationalized as a continuous variable. ‘There are many advantages, both technical and conceptual, to a definition, which identifies an axis of variation rather than mainly distinguishing two or more categories. Not only may one measure differences more finely, and employ a wider range of statistical tools, but one is in a vastly better position to understand the process of change, if one has been liberated from the restricted mind-set of typologies and instead defined many concept in terms of continuous variables’.

These two concepts “conventional religious tradition” and “tension with the sociocultural environment” are used in definitions of three types of religious organizations: churches, sects and cults:

Definition 56. ‘A church is a conventional religious organization.’

Definition 57. ‘A sect movement is a deviant religious organization with traditional belief and practices.’

Definition 58. ‘A cult movement is a deviant religious organization with novel belief and practices’.

The concept of deviance is defined as ‘departure from the norms of a culture in such a way as to incur the imposition of extraordinary costs of from those who maintain the culture (definition 59). It is further elaborated as equivalent to tension in proposition 142: ‘Tension with the surrounding sociocultural environment is equivalent to subcultural deviance, marked by difference, antagonism and separation’. ‘The high tension group is different from the sociocultural standard, mutually antagonistic toward the dominant group that set that standard, and socially separate from them’ (Stark and Bainbridge 1987: 122).

The other axis of variation is less clearly defined and not included in the formal structure of the theory. Stark and Bainbridge elaborated this concept to a greater in their book ‘The Future of Religion’ (1985: 25-26), where they describe more precisely the novel character of cults: ‘whether domestic or imported, the cult is something new vis-à-vis the other religious bodies of the society in question. If domestic – regardless of how much of the common religious culture it retains – the cult adds to that culture a new revelation or insight justifying the claim that it is different, new, “more advanced”. Imported cults often have little common culture with existing faiths; they may be old in some other society, but they are new and different in the importing society’.

Therefore the main criterion of this axis seems to be ‘inclusion of novel elements’ and ‘difference’ in relation to the conventional religious tradition. This kind of ‘difference’ is, however, also mentioned earlier as one of the three criterions of tension, which may pose some conceptual problems. The difference in relation to the dominant religious tradition is usually greater in the case of cults than in the case of sects, so there are more arguments to include it as the one of the constitutive elements of the novelty axis than tension axis. The openness to novel religious elements different from the dominant religious tradition will therefore define the novelty axis of this typology, whereas in the case of tension axis, the antagonism toward dominant religious group seem to be the most distinctive criterion.

            Religious movements may be located along these two dimensions according to the degree of their openness to innovation and antagonism toward the dominant religious group. Among more conventional groups, close to the high-tension pole of this axis, we may locate sects and close to the other pole – churches. An intermediary position should hold intra-church religious movements (as Catholic Charismatic Renewal Movement), which are more sect like than the churches themselves. Among less conventional and more innovative groups – cult movements will hold positions close to the high-tension pole and audience cults and client cults – positions close to the other end of this axis.

Fig. 1. A fourfold typology of religious organizations 

Innovation (-)

Innovation (+)

 Tension (+)

Sects  

Cult movements

Tension (-) 

Churches and intra-church religious movements-

Audience and client cults (cultic milieu)

 

 

            To illustrate this typology empirically we will use data from the questionnaire study of members of 643 members of various religious movements sponsored by a grant KBN 1HO2E02716 and conducted in the years 2000 and 2001 (Doktór 2002a). The variety of these movements, although not fully representative for the Polish religious panorama, includes all elements of the proposed typology: Catholic New Religious Movements, sects, cult movements and client cults (represented only by Rebirthing). With the exception of the latter category they include both imported and ‘indigenous’ movements, which emerged within the Roman Catholic Church (Oasis), or as a schism to it (Reconciliation, Sword of the Spirit), or as a form of religious innovation (Native Faith, Luciferians). The majority of these movements could be classified according to the religious tradition to which they belong, but in some cases it is not possible and they are labeled here as ‘other’.

Table 1. Classification of movements based on major religious traditions.

Groups

Catholic

Protestant

Hindu

Buddhist

Other

Total

Brahma Kumaris
21
21
Zen
12
12
Baptists
23
23
Khordong

11

11

ISKCON

107

107

Neocatechumenate
45
45
Christians of the Evangelical Faith
18
18
Karma Kagyu
29
29
Pentecostal Church
6
6
Luciferians
11
11
Chaitanya Mission
60
60
Sword of the Spirit
17
17
Unification Church
33
33
Mission of Grace
30
30
Oasis
18
18
Reconciliation
8
8
Triangles of Light
8
8
Rebirthing
20
20
Rosicrucians
11
11
Native Faith
10
10
Jehovah’s Witnesses
87
87
Taize
10
10
Bahai
19
19
Water of Life
20
20
Total
73
209
188
52
112
634
 

As we can see in table 1., the most numerous are movement belonging to Protestant tradition both mainline and marginal (Jehovah’s Witnesses as the largest group in our sample). Almost equal in size are Hindu groups with ISKCON as the largest group in the sample. The third in number are movements outside the major religious traditions with Unification Church as the most numerous. Catholic and Buddhist movements constitute the smallest category in our ample with Neocathenumenal Way and Karma Kagyu as the most numerous.

            As an indicator of religious innovation (openness to the inclusion of novel religious elements different from the dominant religious tradition) we will use scores on an item measured on the seven points Likert scale (1 - strongly disagree to 7 - strongly agree):‘Even if people belong to a particular religion, they should still feel free to draw on teachings from other religious traditions’.

Table 2. ‘Even if people belong to a particular religion, they should still feel free to draw on teachings from other religious traditions’ (means and standard deviations of the seven points scale).

Groups

Means

SD

Brahma Kumaris
6,70
,73
Zen
6,58
,90
Baptists
2,74
2,05
Khordong
6,55
,93
ISKCON
6,41
1,19
Neocatechumenate
4,16
1,94
Christians of the Evangelical Faith
3,88
2,60
Karma Kagyu
5,38
1,78
Pentecostal Church
1,60
1,34
Luciferians
6,09
1,22
Chaitanya Mission
6,18
1,22
Sword of the Spirit
3,81
2,46
Unification Church
6,88
,33
Mission of Grace
1,47
1,36
Oasis
3,76
1,99
Reconciliation
4,63
1,60
Triangles of Light
6,75
,71
Rebirthing
5,80
1,47
Rosicrucians
6,64
,67
Native Faith
3,90
1,45
Jehovah’s Witnesses
3,35
2,21
Taize
2,33
1,32
Bahai
7,00
,00
Water of Life
3,37
2,43
Catholic
3,83
1,96
Protestant
3,09
2,25
Hindu
6,39
1,16
Buddhist
5,90
1,56
Other
6,32
1,25
Total
4,94
2,30
 

As we can see from table 2., the mean scores on this item are the highest In the case of cults (Hindu, Buddhist and movements outsider the major traditions). In a study on moral and religious pluralism (RAMP) where this item was used in a research conducted on a Polish representative sample the mean score on this item was 5,28 and was smaller than in the case of cult members and higher than in the case of sect and Catholic movements members (Borowik and Doktór 2001).

The measurement of tension as an antagonism towards the dominant religious group is more complicated. We have used here an index composed of answers to the question ‘Are, according to your opinion, members of your religious/spiritual group different from Catholics? Please, describe both groups on adjectives given below.’ Respondents rated the image of members and Catholics on 7 paired adjective 5–point scales (tolerant/intolerant, educated/uneducated, friendly/unfriendly, not obedient to their leaders/obedient to their leaders, honest/dishonest, creating a threat/do not creating a threat to society, superstitious/non-superstitious). For the sake of brevity we will present here the results for traditions only. 

Table 3. ‘Are, according to your opinion, members of your religious/spiritual group different from Catholics? Please, describe both groups on traits given below. Please choose those, which are most close to your opinion on the scale ranging from 1 to 5’ (mean values of the five points scale).

Tradition
Intolerant
Uneducated
Unfriendly 
Not obedient to their leaders
Dishonest
Do not creating a threat to society
 Non-superstitious

 

Catholics

We

Catholics

We

Catholics

We

Catholics

We

Catholics

We

Catholics

We

Catholics

We

Catholic

3,0

2,3

2,8

2,5

2,7

2,1

3,2

2,7

2,8

2,1

4,3

4,5

2,9

3,6

Protestant

3,9

1,9

3,0

2,5

3,4

1,5

2,9

2,4

3,6

1,7

3,2

4,6

1,8

4,5

Hindu

3,9

2,1

2,9

2,6

3,1

1,9

3,2

2,4

3,1

2,0

3,3

4,4

2,6

3,7

Buddhist

3,8

2,0

2,8

2,3

2,8

2,1

2,4

2,6

2,9

2,1

3,5

4,3

2,2

3,7

Other

3,9

2,0

2,9

2,2

3,0

2,0

2,4

3,0

3,1

2,1

3,4

4,3

2,4

3,8

Total

3,8

2,0

2,9

2,5

3,1

1,8

2,9

2,6

3,2

1,9

3,4

4,4

2,2

4,0

             In the case of all seven adjectives we may observe more positive perception of the own group in comparison to the dominant religious group – the Catholics. The Catholics are perceived to be less tolerant, educated, friendly, honest, obedient to their leaders (with the exception of Buddhist and movements outside major traditions), creating greater threat to society and more superstitious. This more negative image was visible also among members of Catholic Religious Movements.

            Similar question (’When we hear that a person is a member of a particular religious group, we have certain associations and imaginations about him. Below, there are several adjectives describing members of different religious groups. Please choose those, which are most close to your opinion on the scale ranging from 1 to 5’) were asked in a research conducted on a representative sample (Doktór 2002b) the results were slightly different. Catholics perceived themselves more positively (intolerant – 2,9 in comparison to 3,8 in the present study, uneducated – 2,8 in comparison to 3,0 in the present study, unfriendly – 2,7 in comparison to 3,1 in the present study, not obedient to their leaders – 2,9 in comparison to 2,9 in the present study, dishonest – 2,9 in comparison to 3,3 in the present study, do not creating a threat to society – 4,1 in comparison to 3,4, non-superstitious – 3,1 in comparison to 2,3). Only in the case of obedience to the leaders, which may be valued ambivalently, we don’t see any difference. Catholics perceive other groups usually more negatively than themselves but there are also some exceptions. Jehovah’s Witnesses are perceived rather positively. They are seen as being more tolerant, educated, friendly and honest than Catholics. Antagonism is therefore often mutual but usually greater among members of minority groups.

After the reversal of the direction of coding of the last two traits (differently coded than others) and standardization of scores we may create an antagonism index as a mean difference of perception of the own group and the dominant group. The reliability coefficient is acceptable (Cronbach's Alpha =,71).

Table 4. Differences in the perception of the own religious group and Catholics (means of standardized scores and standard deviations)

Groups

Mean

SD

Brahma Kumaris
,26
,26
Zen
-,42
,85
Baptists
,08
,80
Khordong
-,87
1,05
ISKCON
-,19
,61
Neocatechumenate
-,47
,38
Christians of the Evangelical Faith.
,18
,54
Karma Kagyu
,07
,66
Pentecostal Church
,52
,37
Luciferians
,63
,65
Chaitanya Mission
,06
,34
Sword of the Spirit
,59
,33
Unification Church
-,08
,35
Mission of Grace
,18
,37
Oasis
-,62
,50
Reconciliation
,59
,51
Triangles of Light
,35
,66
Rebirthing
-,45
1,16
Rosicrucians
-,02
,74
Native Faith
-,73
,46
Jehovah’s Witnesses
,23
,63
Taize
-,43
,33
Bahai
-,05
,88
Water of Life
,29
,58
Catholic
-,50
,40
Protestant
,26
,58
Hindu
-,12
,61
Buddhist
-,19
,84
Other
-,05
,74
Total
-,01
,68

            The most pronounced antagonism toward the dominant religious group may be observed among members of protestant sects and the least – among members of Catholic Movements. All sect members display high level of antagonism (the highest among members of two groups, which emerged as a result of schism among members of two former intrachurch communities “Reconciliation” and ‘Sword of the Spirit’). Cults are more diversified. The least antagonism is observed among members of Buddhist and Hindu groups, where (with the exception of Brahma Kumaris and Karma Kagyu) mean scores are below zero (a mean point for all movements). The most diversified are movements outside major traditions with members of neopagan movement “native Faith’ scoring as the as the lowest and Luciferians as the highest in this category.

            Both dimensions are relatively independent. Pearson correlation is low and insignificant (-,08 on the individual level and -,15 in the case of data aggregated on the group level). We may now present the empirical localization of each group in this two-dimensional space.

Figure 2. Localization of religious movements in two-dimensional space of tension and innovation.

Brch. K. –Brahma Kumaris

moon – Unification Church

ZEN – Zen

mł – Mission of Grace

bap – Baptists

oazy – Oasis

bk – Khordong

poj – Reconciliation

hk – ISKCON

ptś – Triangles of Light

k – Neocatechumenate

rb – Rebirthing

kcwe – Christians of the Evangelical Faith.

roz – Rosicrucians

kk – Karma Kagyu

rw – Native Faith

kz – Pentecostal Church

sj – Jehovah’s Witnesses

luc – Luciferians

taize – Taize

mcz – Chaitanya Mission

wb – Bahai

md – Sword of the Spirit

wz – Water of Life

             The localization of particular movements in two-dimensional space is principally in accordance with theoretical predictions of the typology. Catholic movements are localized in the lower left quadrant of the low tension and low innovation, sects are localized in the upper left quadrant of the high tension and low innovation and the majority of cult movements in the upper right quadrant of high tension and high innovation. Three of them (Zen, Khordong and Native Faith) display, however, too low level of tension similar to client cults (as In the case of rebirthing) and one (Native Faith) – also too low level of innovation (similar to Catholic movements). We could attribute this anomaly to small number of members of these movements in our sample (it ranges from 10 to 12 so it might be the measurement level). All more adequately represented movements in terms of their size in the sample fit better to our typology especially the largest movements in the sample for each category (ISKCON, Jehovah’s Witnesses, Neocatechumenal Way and Rebirthing). The most clearly delineated are sects, Catholic Movements and client cults (with only one representant in this category). More problematic is, however, the localization of some cults along the antagonism dimension, where three cases do not fit theoretical predictions based on the typology. Differences in antagonism towards the dominant religious group (ordinary Catholics) between lower tension groups (catholic movements and client cults) and higher tension groups (sects and cult movements) are highly significant (F=35,1, p<,000). Even greater are differences in the acceptance of religious inclusivism between low innovation groups (Catholic movements and sects) and high innovation groups (cults) (F= 436,8, p<000). Innovation operationalized in the form of religious inclusivism performs therefore better as factor differentiating between types than tension operationalized as an antagonism towards the dominant group.

            The usefulness of this typology could be tested by applying it to the propositions of Stark-Bainbridge theory of religion concerning the differential recruitment to various types of movements. Education, religious and social capital should have different effects on affiliation with different groups.

According to this theory, participation in higher tension groups is related to less favorable social position on a measures like socioeconomic status or education. The opposite prediction is formulated in relation to participation in innovatory groups. Cults are more appealing for better educated. ‘The real recruiting ground for cults is now among the relatively privileged – or at least among those with sufficient material and cultural resources to permit them to explore new styles of life and faith. Indeed, this may have been the case in America many years ago, for, by their very definition, cults are experiments in cultural innovation. In many spheres of endeavor, the most educated classes promote cultural innovation.’ (Stark and Bainbridge, 1985: 395-6). Therefore, in case of sects we have two arguments for lower level of education among their members (related to compensation of less favorable social position and low level of innovation) so we may expect the lowest level of education among members of this type of religious movements. In case of cults we have two conflicting predictions, but we may expect here higher level of education than in the case of sects, but lower than in the case of client cults and Catholics movements.

The concept of ‘religious capital’, was introduced in a book authored by Rodney Stark and Roger Finke (2000: 120) and defined as ‘the degree of mastery and attachment to a particular religious culture’. The elements of this conceptualization are, however, also present in the theory of religion proposed by Stark and Bainbridge ( as ‘stakes in conformity’), but in the new theoretical approach they are more ‘elegantly’ formulated. According to Stark and Finke ‘In making religious choices [such as conversion or reaffiliation], people will attempt to conserve their religious capital’ (Stark and Finke 2000: 121). We may therefore expect differences in the religious background manifesting in their ‘religious capital’ according to the type of group to which they belong. In the new version of the market theory of religion, the innovation dimension of the typology of religious groups has been skipped and only the ‘tension’ dimension was left. On the basis of this approach we may expect the lower level of religious capital among the members of the high-tension group, whereas on the basis of the older approach we may predict that cult members will have the lowest level of religious capital before joining the movement (in comparison to members of sects and Catholic movements).

            Parallely to the notion of religious capital, Stark and Finke (2000: 119) use the notion of social capital as consisting of interpersonal attachments and formulate the proposition that people in making religious choices will attempt to conserve their social capital. We may therefore expect that the level of social capital among members of high tension and high innovation groups will be lower than in the case of low tension and low innovation groups. As indicators of social capital we will use three measures: percentage of members unmarried at the age of 30 years and over, percentage of members from families in which fathers was absent at least one year before the age of twenty and percentage of members who change the type of their residence in comparison with childhood. 

Table 5. Education, religious capital (percent of regularly practicing mothers, fathers and respondent themselves before the age of 14) and social capital (percentage of members unmarried at the age of 30 years and over, percentage of members from families in which fathers was absent at least one year before the age of twenty and percentage of members who change the type of their residence in comparison with childhood) among members of different types of religious movements.

Type of movement

Academic education

Regular religious practices - mother

Regular religious practices - father

Regular religious practices - respondent in childhood

Mean percentage of regular religious practices in the family

Percentage of unmarried over 30

Percentage of absent fathers in childhood

Percentage of change of the type of residence

%

%

%

%

%

%

%

%
Catholic movements

24,7

71,2

54,8

67,1

64,4

11,0

16,4

24,7

Sects
18,0
47,4
28,2
34,0
36,5
14,4
22,4
42,1
Cult movements
35,2
34,6
22,6
34,6
30,6
22,6
25,3
42,2
Client cults
30,0

60,0

25,0

60,0

48,3

15,0

00,0

35,0

Total
28,1
43,8
28,2
39,0
37,0
18,3
22,6
39,9
 

The results related to the education level of members confirm expectations based on the two-dimensional version of the typology. We may observe that the innovation dimension discriminates better than tension between four types of movements. Differences in the education level between members of low and high innovation groups are statistically significant and not significant in the case of tension. When data are aggregated at a group level (N=25), the correlation of innovatory dimension (inclusivism) with the academic education of members is ,43 (p<,03), whereas in the case of tension (antagonism) is low (,04) and insignificant. ‘Intellectual capital’ in the form of education is therefore significantly positively related to the ‘cultic dimension’ of our typology as predicts Stark and Bainbridge theory of religion, but in the case of tension we do not observe any significant relationship.

The level of religious capital in the families the future converts was the lowest in the case of cult members on all measures employed except percent of regular religious practices among future members in childhood, where it is almost equal among members of sects and cult movements. Cult movement members have the lowest level of regular religious practices in their families, which better fits to the prediction of the older version of the theory. Generally, however, the tension dimension of this typology discriminates slightly better than innovation dimension. The mean percentage of regular practices in the family is 44% in the case of low innovation groups (sects and Catholic movements) and 32% in the case of high innovation groups), whereas in the case of high tension groups (sects and cult movements) it is 33% in comparison to 61% in low tension groups (Catholic movements and client cults) with F- values 16 and 45 respectively.

            In all measures employed members of cult movements have the lowest level of social capital as measured by the rate of interpersonal attachments. Generally, however it is the tension dimension, which best discriminates between the types of movements. Differences on all three measures are statistically significant between high and low tension groups, whereas in the case of innovation it is only the percentage of unmarried over 30, which differentiates significantly between low and high innovation groups.

            As the measures, we have used, are often intercorrelated to evaluate the influence of each one we will apply the multiple regression analysis.

Table 6. Multiple regression with membership in high tension and high innovation groups as dependent variables.

 

Tension

Innovation 

Beta

Sign.

Beta

Sign.

(Constant)
,000
,000
Age
,058
,172
-,141
,001
Sex (female)
-,026
,496
-,073
,060
Regular religious practices in the family
-,254
,000
-,154
,000
Academic education
,028
,476
,193
,000
Unmarried over 30
,030
,474
,118
,005
Father absence
,068
,080
,028
,472
Change of residence
,105
,008
,062
,115
Rsq.
,091
.
,085
.

            The significant predictors of membership in high-tension groups (sects and cult movements) are less regular religious practices in the families of members and more frequent residence change (the rate of father absence is approaching the level of statistical significance). In the case of membership in high innovation groups (cults) we have more predictors: higher level of education, less religious family background, younger age and greater frequency of being unmarried over 30 years.

The role of religious capital is influential in participation in both high tension and high innovation groups, although it is stronger the first case. The role of social capital is not so significant and is manifesting in different indicators (in the case of participation in the high tension groups it is the rate of residence change and in the case of high innovation groups – the rate of unmarried over 30). The role of ‘intellectual capital’ is influential only in the case of participation in high innovation groups (cults), where it is the most influential predictor.

            The results confirm the usefulness of the fourfold typology of religious movements. The differences between the various types could be more adequately explained by two dimensions of modified version of Stark and Bainbridge typology than by one dimension, proposed in the Stark and Finke model, especially in the case of the education level of cult movement members. Their high level of education, noted in many empirical studies, including Stark and Bainbridge (1985) or Bader and Demaris (1996), seem to contradict expectations concerning the characteristics of members of high-tension groups based on other versions of the market model.

The operationalization proposed here includes, however, only one indicator for each dimension. To test this model in a more rigorous way we may need more indicators and greater variety of movements (especially in the category of cultic milieu, which was underrepresented in our study).

References

BADER C., DEMARIS A. (1996) A test of the Stark-Bainbridge theory of affiliation with religious cults and sects. Journal for the Scientific Study of Religion 35(3): 285-303.

BOROWIK I., DOKTÓR T. (2001) Pluralizm religijny i moralny w Polsce. Kraków: Nomos. 

DOKTÓR T. (2002) Innowacje religijne: ruchy, uczestnicy, reakcje społeczne. Olsztyn: Mantis.

 DOKTÓR T. (2002) Factors Influencing Hostility Towards Controversial Religious Groups, Social Compass 4: 553-562.

STARK R., BAINBRIDGE W.S. (1985) The Future of Religion. Berkeley University of California Press.

STARK R., BAINBRIDGE W.S. (1987) Theory of Religion. New York: Lang.

STARK R., FINKE R. (2000) Acts of Faith. Berkeley: University of California Press.


[Home Page] [Cos'è il CESNUR] [Biblioteca del CESNUR] [Testi e documenti] [Libri] [Convegni]

[Home Page] [About CESNUR] [CESNUR Library] [Texts & Documents] [Book Reviews] [Conferences]