The British Journal of Sociology 2007 Volume 58 Issue 4 Religious individualization or secularization? Testing hypotheses of religious change - the case of Eastern and Western Germany Detlef Pollack and Gert Pickel Abstract The individualization thesis advanced by sociologists of religion such as Grace Davie, Daniele Hervieu-Leger, Michael Kriiggeler, Thomas Luckmann, Hubert Knoblauch, Wade Clark Roof, Wayne E. Baker, and others has become increasingly widespread especially in Europe within the sociology of religion. In contrast to the secularization theory it assumes that processes of modernization will not lead to a decline in the social significance of religion, but rather to a change in its social forms. According to the individualization theory, traditional and institutionalized forms of religiosity will be increasingly replaced by more subjective ones detached form church, individually chosen, and syncretistic in character. The article examines the empirical applicability of the individualization thesis on the basis of how religiosity and church affiliation have evolved in Germany over the past 50 years. It comes to the conclusion that the rise of individually determined non-church religiosity cannot compensate for the losses of institutionalized religiosity, since non-church religiosity remains rather marginal and is interwoven with traditional Christian religiosity. Religious individualization is only a component of the predominant secularization process. Keywords: Religiousness; church adherence; non-church religiousness; secularization; individualization; Germany In the course of the past two decades, the ongoing debate between the adherents of the secularization theory (Wilson 1982; Dobbelaere 2002; Bruce 1999, 2002; Norris and Inglehart 2004) and the proponents of the economic market model (Stark and Bainbridge 1985; Iannaccone 1991,1992; Stark and Iannac-cone 1994; Stark and Finke 2000) has attracted much attention in the field of the sociology of religion (Bruce 1992; Young 1997; Swatos and Olson 2000; Jelen 2002). A third approach, however, which clearly distinguishes itself from Pollack and Pickel (European University Viadrina) (Corresponding author email: pollack@euv-frankfurt-o.de) © London School of Economics and Political Science 2007 ISSN 0007-1315 print/1468-4446 online. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Maiden, MA 02148, USA on behalf of the LSE. DOI: 10.1111/J.1468-4446.2007.00168.X 604 Detlef Pollack and Gert Pickel both of these approaches, has remained largely unnoticed: a view that we will call the religious individualization thesis. This thesis has become increasingly widespread especially in Europe within the sociology of religion and, instead of the market theory, which is more prominent in North America, has established itself as the chief adversary of the secularization theory. This theory has been advanced by Grace Davie (1994), Paul Heelas and Linda Woodhead (2005), Daniele Hervieu-Leger (1999,2000), Roberto Cipriani (1989), Michael Kriiggeler and Peter Voll (1993), Hubert Knoblauch (1999) in Europe, but by Wade Clark Roof (1993, 2001), Robert Wuthnow (1998), Robert C. Fuller (2002), Ronald Inglehart (Inglehart and Baker 2000) and others in the USA as well, to name only a few. While the secularization theory assumes a negative correlation between processes of modernization and the vitality of religion, the religious individualization thesis maintains that processes of modernization will not lead to a decline in the social significance of religion, but rather to a change in its social forms. The proponents of the individualization thesis concede that the traditional churches and church-related behaviours in modern societies have been affected by an undisputable decline. However, they contend that this does not indicate a loss of religiousness for the individual. On the contrary -the decline of the established religious institutions goes hand in hand with a rise in individual religiosity. The individualization thesis utilizes a distinction between church and religion in order to distance itself from the secularization theory. In this view, individuals are increasingly freeing themselves from institutional guidelines in their religious ideas and behaviours, and thus increasingly making their own decisions about their religion. In consequence, ever more subjective forms of religion are purportedly replacing institutionalized ones. Davie (1994, 2002), for example, differentiates between traditional religious practices and belonging on the one hand, and religious belief and religious sensitivity on the other. She assumes that 'religious belief is inversely rather than directly related to belonging'. Therefore, 'as the institutional disciplines decline, belief not only persists, but becomes increasingly personal, detached and heterogeneous' and shows a 'reverse' tendency (Davie 2002: 8). 'Believing without belonging' will become the dominant feature of European religiosity, especially in the Northern countries (Davie 1994).1 Hervieu-Leger (1999) likewise observes an increasing loss of control of the established religious institutions over the belief systems and religious practices of individuals. Hence, a gap between the official forms of religion and individually accepted religious perceptions and behaviours emerges. Individuals are more and more freed from established religious authorities and thus enabled to autonomously determine their belief systems, often synthesizing various religious traditions. This process of religious individualization does not mean that individual religiosity is weakened. Instead, it becomes © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 605 multifaceted, syncretistic, and alienated from church. Wuthnow (1998) argues that the decline of organized religion in the USA is accompanied by a rise in spiritual concerns, leading to a shift from a 'spirituality of dwellings', which emphasizes sacred places, to a 'spirituality of seeking', which emphasizes the personal quest for new spiritual paths. Finally, Roof (2001) deduces the erosion of collective engagement with religion in the USA among the younger generation from the declining status and authority of traditional church institutions, the individualization of the quest for spirituality, and the rise of multiple New Age movements including alternative spiritual practices such as astrology, meditation, and alternative therapies. This leads to a diverse bricolage of personal beliefs. Although the individualization thesis has lately become widely acknowledged, it is certainly not a new strand within the sociology of religion. Thomas Luckmann (1967) was the first to advocate this thesis during the 1960s. Since then, it has become a widespread pattern of argumentation in the academic field and evolved into a broadly accepted hypothesis for an increasing number of scholars - whether they directly refer to Luckmann's original approach or not. The aim of this paper is to examine possibilities for the empirical verification or falsification of this approach. Before we can begin the empirical assessment of the individualization thesis (see section 4), however, we must first briefly describe this theory's main lines of argumentation in order to gain a deeper understanding of its underlying assumptions (see section 1). We shall then develop the pivotal questions of our investigation (see section 2) and finally outline our empirical instruments (see section 3). The main result of our analysis will be that the empirical evidence indicates a slight trend towards individualization, but that this effect is quite minute. Thus, it should be regarded as no more than a corollary of a very strong process of secularization in Germany rather than as a compensation for this process. Surprisingly, only few attempts to expose the individualization thesis to empirical verification or falsification have so far been made. The most important test was carried out by Gill (1999) and Voas and Crockett (2005) who, however, only dealt with the case of Great Britain. 1. Thomas Luckmann's privatization theory Luckmann (1967,1991) bases his views on the history of religion on a differentiation between worldview, social structure, and the individual. While the congruence between these three levels is relatively high in archaic societies (Luckmann 1972: 9), advanced civilizations produce an institutionally specialized sphere of religion relatively distinct from their universal worldview. The religious sphere in modern societies reveals only weak connections to this sacred universe. Thus, the institutional specialization of religion in modern British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 606 Detlef Pollack and Gert Pickel societies that Luckmann considers to be a consequence of the functional differentiation of society entails that norms embedded in social institutions are no longer perceived as subjectively significant by individuals, and thereby loose their effectiveness as models for the integration of sense (Luckmann 1967: 85ff.). In other words, institutionally specialized norms are no longer capable of representing the whole of the sacred cosmos, and even the norms conveyed by the Church cannot interpret the world as a whole anymore, but rather become 'restricted to a specifically "religious sphere" ' (1967: 95). Since social institutions, among them the Church (1967:103), are no longer embedded within the sacred cosmos of the worldview, they lose their ability to provide a 'subjectively meaningful system of "ultimate" significance' for individuals (1967:101). This is Luckmann's main argument in explaining the privatization of religion in modern societies. In making sense of the world and his personal life, the individual can no longer depend on institutions to contribute to the formation of his consciousness and personality. He must instead 'select certain religious themes from the available assortment and build them into a somewhat precarious private system of "ultimate" significance' (1967:102) 'in a relatively autonomous fashion' (1967: 98). The 'official' model of religion becomes neutralized as 'mere rhetoric' (1967: 95), while the individual may choose his religiosity 'from a variety of themes of "ultimate" significance' (1967:105), from a multitude of esoteric and religious traditions, e.g. 'psychoanalysis, popular discussions of Marxism, LSD-cults, Zen, etc' (Luckmann 1972:12). This gives individual belief systems, described by Luckmann (1967: 91) as a 'new social form of religion' or as an 'invisible religion', a rather syncretistic character. Thus, the decline in the social relevance of church does not coincide with a general decline in religion or mass secularization (1972: 11). Individuals remain religious, although the new social form of this religiosity takes on a diffuse, private character and no longer claims the status of primary, universally accepted institution (1967: 103). Secularization theories are, in Luckmann's opinion, a 'faulty design' (1991: 179). The falsification of the secularization theory is undoubtedly at the core of Luckmann's approach within the sociology of religion.2 In order to prove his claim, he emphasizes the discrepancy between subjective religiosity and official religious models. It is this distinction that allows him to concede the losses of traditional forms of religion, while at the same time contending the continuing subjective significance of religion. This shift from public to private is identified as the privatization and individualization of religion. The individualization thesis as firstly elaborated by Thomas Luckmann has meanwhile been taken up by an increasing number of sociologists and attained a certain degree of prominence in the sociology of religion. Although this approach has become a generally accepted pattern of argumentation in the sociology of religion, it has also often been criticized for several reasons: © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 607 1. Scholars have repeatedly criticized the imprecise use Luckmann makes of the term 'religion'. Nearly every sociologist who interprets Luckmann (see for example the discussion in Hamilton 2001:184) points this out. In this context, it is also often stressed that Luckmann's refutation of the secularization theory is already implicit in his broad use of the term religion (Tyrell 1996: 445). These criticisms tend to go hand in hand with attempts to formulate more precisely what Luckmann means by his 'new social form of religion' or his 'invisible religion'. Hubert Knoblauch (1991:22) coined the 'search for the invisible religion' as a path on which the sociology of religion is currently headed. Many sociologists of religion explain this term by pointing to the currently popular New Age movement or the adherence to non-Christian forms of spirituality such as Zen meditation, belief in reincarnation, astrology, the esoteric sector, or new interest in psycho-cults, body culture, and group therapy (Honer 1985; Mörth 1989; Zinser 1997; see also Knoblauch 1991:28,31; 1999:127). Most scholars are aware of the fact that the so-called 'invisible religion' encompasses diverse phenomena that are difficult to analyse as a common field of beliefs and practices (Wohlrab-Sahr and Krüggeier 2000: 241). However, if the individualization thesis is not to remain purely tautological, the broad term 'invisible religion' must be defined more precisely in order to render it testable. 2. In contrast to his concept of religion, most scholars tend to agree with Luckmann's assertion of a setback in traditional church adherence and a subsequent rise in subjective, more or less syncretistic, non-traditional religiosity. There is a vast quantity of statements noting a decline in institutionalized social forms of religion and a rise in extra-institutional forms of religion (see Machalek and Martin 1976; Davie 1994; Voll and Krüggeier 1992; Ebertz 1997; Wuthnow 1998; Hervieu-Leger 1999; Zulehn er, Hager and Polak 2001; Heelas and Woodhead 2005). None the less, critical assessments of this aspect of Luckmann's theory are also to be found. For example, Karl Gabriel criticizes his overstressing of the social marginalization of churches. He argues that the 'invisible, extra-institutional', privately produced religion is not the 'actual' place of religion in modern societies (Gabriel 1992: 20). Instead, Gabriel assumes that the institutionalized social form of religion still exert great influence on the religious market. This claim is supported by the Church's monopoly on rituals, or by the strong role it plays in the public sphere (Gabriel 1993: 29f.). Tyrell (1996), Usarski (1988), Gärtner (2000: 87), and others assert that the extraordinary influence of alternative religious orientations and behaviors is greatly overestimated within the sociology of religion. 3. Luckmann's thesis surrounding the individualization and privatization of religiosity is also largely positively acknowledged in the discussion. British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 608 Detlef Pollack and Gert Pickel Berger (1980), Kriiggeler and Voll (1993), Krech (1998), Hervieu-Leger (1999), Roof (2001), and others maintain that the significance of institutional patterns in the development of religious attitudes and behaviours seems to be declining. They also affirm that individuals are composing their religious orientations more and more independently from a wide range of religious traditions. Voll (1993:241) argues that in the process of structural change within the social spheres of modern societies, religion has largely lost its structural contours. However, some scholars raise doubts concerning the privatization and individualization concept. Jose Casanova (1994) entirely refutes the idea that 'privatization' is an adequate term for describing the developments of religion in modern societies. He in turn stresses the increased influence of religion on the public spheres of modern societies. 2. Questions guiding our inquiry If we are to test the empirical applicability of the individualization thesis and take into account the critical assessments outlined above, we must deal with the following questions: 1. Has traditional church affiliation really lost in significance in the course of the socio-structural reorganization of modern societies, or does the individualization thesis, as Gabriel argues, underestimate the social value of the official model of religion? 2. How has individual religiosity evolved in comparison to traditional church affiliation? Has it also lost in significance, as could be contended, or, as the proponents of the individualization thesis claim, has its evolution diverged from that of institutionalized religion? 3. Can any valid statements be made concerning the new social form of religion if one tries to limit this form to empirically measurable phenomena such as psycho-cults, esoterism, occultism, magic, astrology, Zen, or self-realization therapies, as the literature since Luckmann has suggested? Has the significance of such forms of non-church religiosity increased? 4. What is the precise relationship between this individual or new religiosity and traditional church affiliation! Is the former replacing the latter, are the two merging, or do they coexist as alternatives to one another? 5. How can one describe the relationship between the religious dimensions mentioned here and processes of individualization! Is subjectification, as Luckmann assumes, a general trait of an individual's relationship towards religion (its new as well as its traditional forms) or is one form of religion favoured over others through processes of individualization? Last but © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 609 not least, does individualization have any effect at all on interest in religious forms, be they more traditional and institutionalized or subjective and diffuse? This paper attempts to provide answers to these questions. In the following, they will be examined on the basis of how religiosity and church affiliation have evolved in Germany over the past 50 years. As the individualization thesis claims to be of universal relevance, it should apply to all modern societies. Germany, however, provides a particularly interesting case since it can be divided into two 'sub-cases'. West Germany belongs to the modern societies for which the theory was first conceptualized, and can thus be considered as a typical example that should not obviously contradict the theory. East Germany, which was divided from West Germany for 40 years, is still in the process of catching up to the western German modernization level. This case can thus serve to identify developments in the relationship between traditional church affiliation and individual or new religiosity in a society where non-church members and non-believers form the majority of the population. Does the higher degree of secularization foster or hinder processes of religious individualization? The case of East Germany can thus provide valuable insights regarding the direction the relationship between traditional and new forms of religiosity is developing in, if the decline of church affiliation continues. 3. Operationalization In order to empirically measure the development of religiosity and church adherence in Germany, we utilized the concept of religious dimensions provided by Charles Glock (1962). He distinguished five dimensions of religion: the dimension of religious belief; the dimension of religious experience; the cognitive dimension; the dimension of church-ritual practice; and the dimension of ethical consequences. For the purpose of this paper, it seemed sensible to eliminate the ethical and cognitive dimensions, which are relatively independent from the other units (Stark and Glock 1968; Kecskes and Wolf 1993), to integrate the dimensions of belief and experience, since they belong together (Clayton and Gladden 1974; Kecskes and Wolf 1993), and to distinguish these from the church-ritual dimension. Within the dimensions of belief and experience, however, we must make a further distinction between Christian ideas of faith and non-Christian convictions. Whereas in the 1950s and 60s non-institutional, diffuse, and non-Christian forms of religion were often not included in sociological studies on religion, we are particularly interested in these forms and must therefore place special emphasis on them. Thus, we end up with three dimensions in our understanding of religion: (1) traditional British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 610 Detlef Pollack and Gert Pickel church affiliation, (2) individual Christian religiosity, and (3) religiousness outside church or non-church religiosity. In many sociological studies of religion, traditional church affiliation is described on the basis of indicators such as church membership, the quantity of people officially joining or leaving the church, frequency of church attendance, participation in church life, trust in the church, etc. As there exist church statistics and extensive survey data covering a 50-year period for three of these variables3 - church membership, quantities of people officially leaving the church, and frequency of church attendance - these three will primarily be used to measure this religious dimension. There are also several indicators available to measure individual Christian religiosity. The International Social Survey Programme (ISSP) and the World Value Survey (WVS) use belief in God and the self-assessment of being religious as indicators for individual Christian religiosity. Two surveys of the population of western Germany conducted by Emnid at a twenty-five year interval - 1967 and 1992 - included a number of additional items on Christian faith. Respondents provided information on their belief not only in God, but also in life after death, resurrection, virgin birth, original sin, etc. For belief in life after death, the available data even goes back as far as 50 years. On the other hand, the diffuse, open-ended field of non-church religiosity is difficult to define. Since 1982, the International Social Survey Programme (ISSP) has added a package of four questions in order to assess non-church religiosity: whether good luck charms sometimes do bring good luck, whether some fortune tellers really can foresee the future, whether some faith healers do have God-given healing powers, and whether a person's horoscope can affect the course of his or her future. One glance at these variables reveals that they all represent indigenous forms of popular religion and not the 'alien' forms of non-church religiosity, such as various East Asian forms - Zen, New Age, Yoga, Reiki or esoterism - which are of particular interest in our inquiry. In order to gain an impression of these non-indigenous social forms of religion as well, we developed a new instrument including a total of seven questions pertaining to the acceptance of non-church religious forms - both indigenous and East Asian forms of non-Christian religiosity4 - and conducted two representative surveys in western and eastern Germany in 1999 (Gul) and 2000 (PCE), which included this set of questions among others.5 By employing this instrument, we attempted to define Luckmann's concept of 'invisible religion' more precisely and to thus render it empirically testable. This instrument contained questions covering6 - respondents' belief in New Age, - the effectiveness of Zen meditation and Yoga, - reincarnation, © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 611 - magic, spiritualism, occultism, - astrology, - faith-healers or spiritual healers, - mysticism (Sufism/cabbalism). The explorative factor analysis demonstrates that all variables load on one factor (not shown here). A confirmative factor analysis attempting to divide the related variables into two given dimensions reveals, however, that there exist differences in the strength of the correlation between the various forms of non-church religiosity (also not shown here). Indigenous forms of non-church religiosity centre on the astrology, faith healers and, to a lesser extent, reincarnation variables, whereas the East Asian forms clearly involve New Age and Zen. In both cases, other variables are positioned at various distances from these two cores. In order to retain both the differences between the two dimensions and their overlaps, we will work with the factor values for both dimensions in the following. Finally, with regard to the empirical survey of the concept of individualization, the difficulties arising are, as expected, also quite high. On a socio-structural level, we can speak of individualization, as in Jagodzinski and Quandt's work (Jagodzinski and Quandt 1997:763; see also Beck 1992), whenever the differences in religious characteristics within socio-structurally marked groups are larger than those between such groups; in other words -when socio-structural differences have no significant influence on religious characteristics and differences in age, education, income, place of residence in their influence on religious orientations and practices get blurred. On a semantic level, we can speak of individualization when respondents agree with individualization items expressing values such as self-determination, self-realization, enjoyment of life, or freedom in general. General individualistic orientations are shown by agreement with statements such as the following: T would like to lead a somewhat unusual life.' 'It is very important for me to realize my personal aims.' T enjoy convincing other people of my opinion.' as well as by disagreement with the assertion: T try to respect law and order in all circumstances.' These four items allow us to construct an index to test individualization on the semantic level. The question that interests us here is how close a correlation there is between acceptance of the East Asian type of non-church religiosity and the other religious dimensions on the one hand, and agreement with the individualization index on the other. In accordance with Luckmann's (1991: 181) distinction between structural and semantic individualization, the analysis British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 612 Detlef Pollack and Gert Pickel of the influence of individualization on religious convictions and behaviour is carried out on two levels: a structural and a semantic one. 4. Empirical results 4.1. Traditional church affiliation An examination of the development of traditional church affiliation over the past fifty years shows that every indicator taken into consideration points towards a weakened significance of churches. Church membership in western and eastern Germany amounted to over 90 per cent in 1949. In 2000, it had declined to about 80 per cent in the west and about 25 per cent in the east (Gul 1999; PCE 2000). Within 50 years, the share of people not belonging to any church increased more than three times in western Germany and about ten times in the eastern part of Germany. The same trend also applies to the indicators 'numbers of people officially leaving the church' and 'frequency of church attendance'. A closer look at the development of the rate of secession - people officially leaving the church (cf. Figure I) - demonstrates that the curves are remarkably parallel for the Catholic Church and Lutheran churches. This parallelism is graphic evidence of the fact that people leave the church primarily for reasons external to the church itself, and less as a result of church action. Periods of rapid social change such as at the end of the 1960s and the beginning of the 70s, or the time period after re-unification, seem to have a strong impact on the rate of secession. A closer examination of this phenomenon also exhibits that the rate of Lutherans leaving the church remains steadily above that of secessions from the Catholic Church. This corresponds to the general observation that the social bonds of the Catholic Church are stronger than those of the Lutheran churches. In particular, however, the graph shows that the plateaus of the curve rise continuously following individual waves of secessions, and that the rate of secession never falls back to its original level. This indicates that the erosion of the churches is picking up in speed, and that we cannot expect the dynamics of this erosion to decelerate in the near future. If we additionally take into consideration that among the younger cohorts of those leaving church, the highly educated are no longer overrepresented (Engelhardt and Loewenich and Steinacker 1997:317) we must speak of a normalization of church withdrawals and can predict a further dissemination of this phenomenon. These observations are confirmed if we regard our third indicator for traditional church affiliation - attendance at church services. Here again, we must note a dramatic drop over the past fifty years, particularly in the case of the Catholic Church. Whereas surveys for 1952 showed that 51 per cent of Catholics attended church regularly, this number had dropped to 26 per cent by 1999 © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 613 FIGURE I: Withdrawals from church in western Germany 1945-1999 as a percentage of all members 1,2-,-1 — Protestant — Catholic Source: Institut für kirchliche Sozialforschung (IKSE) 1999; Evangelische Kirche in Deutschland (EKD) 1990-96; Kirchenamt der Evangelischen Kirche in Deutschland 2000. TABLE I: Regular church attendance among Catholics and Protestants in western Germany, 1952-1999 (%) Regular church attendance 1952 1963 1967/69 1973 1982 1991 1999 Catholics as a whole 51 55 48 35 32 33 26 Catholics in the age of: 16-29 52 52 40 24 19 17 10 30-44 44 51 42 28 26 21 15 45-59 50 56 53 46 29 34 24 60 and older 63 64 62 57 54 54 50 Protestants as a whole 13 15 10 7 6 8 7 Protestants in the age of: 16-29 12 11 6 3 4 4 4 30-44 7 10 6 3 4 4 3 45-59 13 16 11 7 6 7 6 60 and older 23 24 22 12 12 17 15 Sources: Institut für Demoskopie Allensbach 1978ff., in general n = 1000. (cf. Table I). Differences between age groups were relatively small in 1952 in comparison to 1999, when the share of Catholic Church members over the age of 60 who regularly attended church services was five times higher than church members between the ages of 16 and 29, and Lutherans over the age of 60 who British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 614 Detlef Pollack and Gert Pickel attended church regularly was approximately four times higher than Lutherans aged 16-29. The group of regular churchgoers is thus primarily composed of the elderly. This means that in comparison to the shrinking number of church members this group will, in the future, over-proportionally shrink. 4.2. Individual Christian religiosity We first measured the development of individual religiosity with a relatively broad spectrum of variables, and later concentrated on two - the development of belief in God and the development of belief in life after death. As shown by a comparison of religious convictions in 1967 and 1992, acceptance of religious ideas dropped over this period among the entire western German population (cf. Figure II). The only two exceptions were belief in original sin and belief in life after death. If we verify this claim with another survey measuring belief in God as the central religious conviction, we attain the same result. Not only was the proportion of western Germans claiming never to have believed in God higher in 1998 than in 1991 (cf. Table II). Even more striking is the fact that the proportion of those who claim to have previously believed in God, but now no longer do, is clearly higher than those who claim not to have previously believed in God, but do so now. In other words, significantly more people lost their faith in FIGURE II: Development of religious beliefs in western Germany 1967 and 1992 (%) God exists. The Last Judgement does exist. Original Sin does exist. Hell does exist. There is a life after death. Jesus is the son of God. Jesus awoke the dead. Jesus resurrected. 0 10 20 30 40 50 60 70 80 ■ 1967 H1992 Source: Selected statements regarding religiosity and belief; 'SpiegeP-Studies 1967 and 1992 (n = 2500), 'Der Spiegel' 25/46, June 15,1992. © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 615 TABLE II: Increase and decrease of belief in God, western and eastern Germany, 1991 and 1998 (%) Belief in God West East Germany Germany 1991 1998 1991 1998 Have never believed in God 10 13 50 58 Have previously believed in 23 25 25 17 God but do not do so now Have not believed in God 9 11 5 7 previously but do so now Always believed in God 58 51 20 18 Source: ISSP 1991 (West n = 1346; East n = 1486) and 1998 (West n = 1000; East n = 1006); statements are summing up to 100 per cent. God over the past few decades than began to believe. If individual religiosity is measured on the basis of these indicators, and, more importantly, if a central belief in God is postulated as the main element of this religiosity, we must conclude that, contrary to the assumptions made by Luckmann and his followers, both traditional church affiliation and individual religiosity have diminished. One already noted exception to this trend is belief in life after death. If we observe the development of this indicator over time (cf. Figure III), we can observe a decline in the 1960s and 70s, but an increase in the 80s and 90s. This evolution lends support to the proponents of the individualization theory who assume a diverging development of church affiliation and individual religiosity, and it indicates that correlations are perhaps more complex than initially assumed. 4.3. Non-church religiosity There are virtually no time series available for the evolution of agreement with ideas relating to non-church religiosity. The only question posed in the same form over several decades regards belief in the influence of the stars on human destiny. Here again (cf. Figure III), we can observe the same trend: a decline in the 1960s and 70s, and an upturn in the 80s and 90s. This development can also be viewed as an indication of divergence between church affiliation and religiosity. 4.4. Systematic correlations between various dimensions of religion What, then, is the systematic correlation between traditional church affiliation, individual Christian religiosity, and non-church religiosity, as measured by the above indicators? If we begin by making a simple comparison of the frequency British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 616 Detlef Pollack and Gert Pickel FIGURE III: Belief in life after death in western Germany, 1956-1997 and belief in the influence of the stars on human destiny in western Germany (%), 1950-1995 70 I-1 10 0 '-' 1950 56 63 71 75 80 82 86 92 95 97 "•"Life after death (yes) "*"Belief in a connection between human destiny and stars (yes) + Life after death (no) Belief in a connection between human destiny and stars (no) Source: Noelle-Neumann and Piel 1983: 123f.; Noelle-Neumann and Köcher 1993: 215; Noelle-Neumann and Köcher 1997: 256, 293. TABLE III: Traditional church affiliation, individual Christian religiosity and non-church religiosity in western and eastern Germany in comparison, 1999/2000 (West / East) (%) Total Catholic Protestant Non-denominational Church membership 81/24 Monthly church attendance (minimum) 25/7 40/40 20/23 4/0 Belief in God as a higher being 74/27 84/77 71/79 52/12 Religious self-assessment 54/20 65/92 53/73 25/3 Reincarnation (return after death) 15/9 15/30 12/16 19/5 Astrology/horoscope 16/11 17/17 14/14 12/10 Faith-healer/spiritual healer 6/6 7/29 3/3 9/6 Magic/spiritualism/occultism 6/4 6/21 6/3 10/3 New Age (holistic way of thinking) 12/3 10/11 12/3 16/2 Zen meditation/Yoga 24/16 24/36 22/9 28/16 Mysticism 5/4 4/22 3/4 11/3 Source: Western Germany Gul 1999 (n = 1002); eastern Germany PCE 2000 (n=1026); results from a scale with four possibilities, here 1 (strongly agree) + 2 (agree). distributions, we see that all three dimensions are characterized by a significant difference between eastern and western Germany (cf. Table III). For the overall population, this difference is strongest in the area of traditional church affiliation. Here, the indicators show values that are three times higher for western Germany than those for eastern Germany. In the case of individual © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 617 Christian religiosity, values for the western part of the country are still more than twice as high as those for the eastern part - also an unusually high difference. And although they may not be as important as concerning the other two dimensions, east/west differences are significant even in the area of non-church religiosity. This would indicate that, although there is no identity between the various dimensions of religiosity and church affiliation, there is indeed a close correlation. And yet, the correlation between belonging to a church denomination or not and the three dimensions of religiosity has a different structure. In the area of traditional church affiliation, the differences between those who belong to a denomination and those who do not are particularly pronounced: non-denominationals (people who do not belong to any religious community) seldom go to church, although here again there are observable differences between eastern and western Germany. Differences between church members and non-church members are distinctly less significant with regard to belief in God and religious self-assessment, at least in western Germany, where, after all, half of all non-denominationals believe in God or in a higher being, and a quarter regard themselves as religious. The corresponding figures for eastern Germany are only 12 and 3 per cent, respectively. If we look at non-church religiosity, denominational differences are hardly noticeable, and indeed they even tend to be reversed. Here, values for non-denominationals in the former West Germany are mostly higher than for those who belong to a denomination - a statement which does not apply to eastern Germany, where acceptance of non-church forms of religiosity is highest among Catholics. Whereas traditional church affiliation and Christian religiosity are in some cases clearly less accepted by non-denominational respondents than by church members, non-church religiosity has a slightly higher value among non-denominationals than the other two dimensions of religiosity. Nevertheless, it should be noted that the acceptance rate of non-church forms of religiosity is not particularly high among non-denominationals either. In order to systematically determine the relationships of traditional church affiliation, individual Christian religiosity, and non-church religiosity to each other, we made a separate correlation analysis for eastern and western Germany based on the indicators developed above. With the help of these indicators, we examined the degree of correlation between the individual religious dimensions (cf. Table IV). This analysis shows a high correlation between traditional church affiliation and individual Christian religiosity (cf. for instance, the correlation coefficients of r = 0.26 for the relationship between church attendance and belief in God, or r = 0.49 for the relationship between church attendance and religious self-assessment in western Germany, see Table IV). In addition, there seems to be a clear connection between individual Christian religiosity and non-church religiosity (e.g. r = 0.21 between belief in God and non-church religiosity - indigenous, and r = 0.18 between belief in British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 618 Detlef Pollack and Gert Pickel God and non-church religiosity - East Asian in western Germany). On the other hand, there is no evident correlation between traditional church affiliation and non-church religiosity in western Germany. Of course, this also means that there is no negative correlation between these two dimensions either: non-church religiosity does not replace traditional church affiliation. It cannot compensate losses in traditional church affiliation, but it is also not supported by them. In eastern Germany, correlations between the individual religious dimensions are much stronger. Between church membership and belief in God, for instance, the correlation is not a mere r = 0.22, as in western Germany, but r = 0.80 (cf. Table IV); between church attendance and belief in God, the correlation is not merely r = 0.26 but r = 0.46. This closer connection between the various dimensions of religiosity is also evident in light of the fact that the correlation between traditional church affiliation (church membership and church attendance) and the indigenous forms of non-church religiosity is not insignificant, but rather positive. In contrast, traditional church affiliation and East Asian non-church religiosity show no correlation in eastern Germany as well.7 Although indigenous and East Asian non-church religiosity load on one factor in the explanatory factor analysis,8 one sees that both factors are in different co-relational structures to other variables, which once again retroactively justifies the differentiation between indigenous and East Asian non-church religiosity. A comparison of the results in Table IV shows that the high degree of disaffection with the church and secularization in eastern Germany seems to be pushing the various religious dimensions towards convergence. Religion and church are obviously so irrelevant in many social strata in eastern Germany that all forms of religion are rejected, not only traditional church affiliation. Conversely, traditional church affiliation strongly attracts all other religious dimensions (with the exception of East Asian non-church religiosity) and there is a relatively close correlation between them. In western Germany, TABLE IV: Correlations between traditional church affiliation, individual Christian religiosity and non-church religiosity in western and eastern Germany, 1999/2000 (West/East) Church Believe Religious Non-church Non-church attendance in God self-assessment religiosity - religiosity - indigenous East Asian Church membership 0.21/0.47 0.22/0.80 0.30/0.77 n.s./0.24 n.s./n.s. Church attendance 0.26/0.46 0.49/0.52 n.s./0.18 n.s./n.s. Believe in God 0.58/0.83 0.21/0.36 0.18/0.18 Religious self-assessment 0.13/0.32 0.14/0.11 Non-church religiosity - 0.45/0.51 indigenous Source: Western Germany Gul 1999 (n = 1002); eastern Germany PCE 2000 (n = 1026); all values pearsons product-moment correlations, n.s. = not significant on level p < 0.05. © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 619 however, where religion and church enjoy broader social acceptance, the overall religious field exhibits a greater degree of internal plurality. There also, traditional church affiliation and individual Christian religiosity are closely interwoven, but correlations to non-church religiosity are lower. In other words, while religiosity and secularity are more strongly polarized in the eastern part of the country, there is no clear-cut difference between these two poles in western Germany. This is confirmed by a comparison of the figures in Table III, which clearly reveals higher religious values for non-denominationals in western than in eastern Germany. While separation from the church does not principally exclude religious orientations in the former West Germany - although it does weaken them, particularly in their traditional dimensions - there are hardly any religious ties at all to be found among non-denominationals in eastern Germany. 4.5. Individualization and dimensions of religiosity If we now examine the correlation between processes of individualization and dimensions of religiosity on the structural level, our first observation is that socio-structural characteristics do retain a certain influence on the extent and form of religiosity after all. The older the respondents, the more they are inclined to favour traditional forms of church affiliation and individual Christian religiosity (cf. Table V). In contrast, non-church religiosity is more popular among younger respondents. To a large extent, this applies to both western and eastern Germany. Gender also has an effect on all dimensions of religiosity, with women tending to be more religious than men. Higher education levels generally tend to have a negative correlation with traditional church affiliation and individual Christian religiosity in both eastern and western Germany, while for non-church religiosity this correlation is more positive. Urban life tends to promote non-church religiosity and to have a rather negative effect on traditional church affiliation. The effects of a religious education on traditional church affiliation and individual Christian religiosity are clearly positive. In the case of the East Asian forms of non-church religiosity, however, religious socialization has no positive effect, and in the west the same applies to the indigenous forms as well. In other words, religious education and the influence of socio-structural characteristics give rise to clear differences between traditional church affiliation and individual Christian religiosity on the one hand, and non-church religiosity on the other. The two former dimensions tend to be affected negatively by modernization effects such as better education or urbanization, while the latter mostly profits from them. This structure repeats itself when we, on the semantic level, compare individualization items with the indices of religiosity. The correlation between traditional religiosity and individualization, and even between individual Christian religiosity and individualization, tends to be negative,9 while British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 TABLE V: Correlation between the various forms of religiosity and socio-structural characteristics and individualization items in western and eastern Germany, 1999/2000 Church Membership Church Believe Religious Non-church Non-church attendance in God self-assessment religiosity - indigenous religiosity - East Asian Western Germany Age n.s. 0.15 0.14 0.18 -0.15 -0.13 Gender 0.12 0.08 0.14 0.13 0.19 0.10 Education n.s. n.s. n.s. -0.10 0.10 n.s. Urban area -0.17 n.s. n.s. n.s. n.s. n.s. Religious upbringing 0.26 0.20 0.29 0.31 n.s. n.s. Extraordinary life n.s. -0.12 n.s. -0.09 n.s. 0.22 Individualization n.s. -0.15 -0.09 -0.15 n.s. 0.18 Eastern Germany Age 0.19 0.09 0.17 0.22 n.s. -0.18 Gender 0.08 n.s. 0.11 0.10 0.13 n.s. Education n.s. -0.06 n.s. n.s. n.s. 0.13 Urban area -0.08 -0.06 n.s. n.s. n.s. 0.18 Religious upbringing 0.58 0.38 0.58 0.64 0.19 n.s. Extraordinary life -0.15 -0.15 -0.09 -0.14 0.20 0.26 Individualization -0.10 -0.16 n.s. -0.10 0.10 0.21 Source: Western Germany Gul 1999 (N = 1002); eastern Germany PCE 2000 (N = 1026); all values pearsons product-moment correlations, n.s. = not significant on level p < 0.05. Religious individualization or secularization 621 between non-church religiosity and individualization it is mostly positive (cf. Table V). There exists a marked difference between eastern and western Germany in this regard: in eastern Germany, the individualization items correlate with both the East Asian and the indigenous forms of non-church religiosity; in western Germany, only with the East Asian. Even though East Asian and indigenous non-church religiosities are closely related in terms of factor analysis, the force of their structural development is obviously different, so that New Age, Zen meditation, mysticism, and spiritualism must be treated as relatively independent religious phenomena. This is further evidenced by the fact that they are not affected by whether respondents enjoyed a religious upbringing in western or in eastern Germany. However, the effects of individualization should not be overestimated: Table III reveals that the overall number of adherents of East Asian non-church religious forms is relatively small.10 This means that even though modernization indicators such as higher education levels, urbanization, and individualization seem to have opposing effects on traditional church affiliation and individual Christian religiosity on the one hand, and non-church religiosity on the other, one should not infer that the trend to stray from institutionalized forms of religion suggests an upswing in the popularity of non-institutional religious forms. The religious segments promoted by modernization processes are simply too insignificant for such an assumption. Yet, it is obvious that, even in view of these rather small numbers, the religious field has become increasingly pluralized: if the processes of modernization reflected by the variables of socio-structural characteristics and the semantics of individualization have a negative effect on the various religious dimensions in one case and a positive effect in the other, this means that the individual religious dimensions are drifting apart and diversifying. Table V measured individualization on the basis of an individualization index11 and agreement with the statement that a respondent would like to lead an unusual life. In order to strengthen a socio-structural definition of individualization in contrast to the semantic understanding of individualization substantiated by this measurement, we will conclude by taking another look at the influence of socio-structural characteristics on the dimensions of religiosity analysed here. While Table V already contained socio-structural variables, the correlations shown there, for instance between the urban/rural difference or education and the various dimensions of religiosity, were quite weak. This could change if types of religious orientation were to crystallize on the basis of a cluster analysis, and then examined as to their dependence on socio-structural factors. Whereas correlation matrices show only linear correlations, cluster analyses can also describe non-linear relationships.12 They go beyond the conventional multivariate analyses generally used to explain social and political factors that influence the social significance of religiosity and church adherence. It thus seems that cluster analyses are more appropriate for British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 622 Detlef Pollack and Gert Pickel grasping the multi-dimensional character of religious orientations and behaviours. Because cluster analyses focus on typologies of people rather than on correlations of variables, they will represent different groups of individuals according to their different styles of religiosity. Six distinct types of religiosity emerge as a statistically stable and plausible result of the cluster analyses. The non-religious group exhibits a clear profile: these respondents reject all forms of religiosity - traditional, individual Christian, and non-church (cf. Table VI). In western Germany, the majority of this group - which represents almost a third of the population - is affiliated with a denomination (69 per cent), while in eastern Germany only 3 per cent of non-religious people (cf. Table VII), who account for roughly two-thirds of the population, are denominationally affiliated. In western Germany, therefore, there are non-religious people both outside and inside the churches and secularization has become a considerable tendency within the churches. In eastern Germany, non-religious people form the vast majority of the population and evidence weaker ties to church than those in the western part of Germany. The average Christians hardly go to church, but they do tend to consider themselves religious and believe in God in above-average numbers (cf. Table VI). Their endorsement of non-church forms of religiosity is hardly higher than that of the population as a whole - in other words rather weak. In contrast to the non-religious group, the proportion of average Christians officially affiliated with a denomination lies above the share of the overall population that belongs to a denomination. But in spite of this - and this applies particularly to the former West Germany - they hardly participate in church life. They go to church only rarely, perhaps only at Christmas or for special family occasions, and do not exhibit high interests in church affairs. However, they are not extraordinarily skeptical towards the church either. Trust in the church equals the average of the population (in the West) or is above average (in the East). Even if average Christians usually do not take part in church meetings and services, high proportions of them belong to the church and identify themselves as being religious and believers. The church-religious group and the committed Christians differ only slightly from one another. A high level of church affiliation and Christian religiosity characterizes both groups. The religious orientation of committed Christians focuses more on participation in church life, while for the church-religious group, it centres more on individual religiosity. Differences, however, are no more than marginal. The syncretists evidence a strong mix of traditional church affiliation, individual Christian religiosity, and non-church religiosity. They account for no more than 10 per cent of the overall population in both eastern and western Germany, and they differ from the non-church religious group, which exists only in western Germany, in that they participate in church life and attend church services in above-average numbers13 - something that the non-church © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) TABLE VI: Religious orientations in western and eastern Germany: results of a cluster analysis Share in % Belief Religious Churchattendance Trust in Non-church Non-church in God self-assessment church religiosity - indigenous religiosity - East Asian Western Germany Areligious people 30.7 - --- --- — - - Average Christians 29.4 + + — / (") (") Committed Christians 9.0 ++ +++ +++ ++ / (") Church-religious people 8.0 +++ +++ ++ ++ / / Syncretists 7.9 ++ ++ + / / (+) Non-church religious people 15.1 ++ + -(-) / ++ ++ Eastern Germany Non-religious people 65.8 --- --- --- - (") / Average Christians 20.0 ++ + "(-) +++ + / (+) Committed Christians 1.5 ++ ++ ++(+) ++ Church-religious people 2.7 ++(+) ++ ++ ++ + / Syncretists I 8.1 ++(+) ++ + ++ + + Syncretists II 1.9 ++(+) ++(+) ++(+) ++ ++ + Source: Western Germany Gul 1999 (n = 1002); eastern Germany PCE 2000 (n = 1026; +++ = clearly above mean;---= clearly below mean; / = on mean value; all other values in between; type of cluster analysis (calculated with SPSS) partitionary hierarchical. TABLE VII: Styles of religiosity and socio-structural characteristics in western and eastern Germany Church membership Education Urban/rural Age Women Religious upbringing Individualization Western Germany Total 84% 54% 75% 50% Non-religious people 69% middle more urban average 44% 57% ++ Average Christians 91% middle more rural average 55% 82% — Committed Christians 97% lower more rural higher 58% 92% --- Church-religious 100% higher more urban higher 66% 87% -(-) Syncretists 97% higher more rural average 67% 92% Non-church religious people 75% higher more urban lower 60% 72% ++ Eastern Germany Total 24% 53% 41% 26% Non-religious people 3% higher average lower 50% 17% + Average Christians 42% lower average higher 59% 69% - Committed Christians 100% very low more rural higher 47% 100% --- Church-religious 96% low average higher 61% 89% — Syncretists I 89% low average higher 55% 85% -(-) Syncretists II 95% low more rural much higher 65% 100% -(-) Source: Western Germany Gul 1999 (n = 1002); eastern Germany PCE 2000 (n = 1026). Religious individualization or secularization 625 religious group does not do. On the other hand, non-church religious respondents approve more strongly of non-church religious forms than the syncretists. For them, religiosity has a clearly non-church, indeed perhaps an anti-church character, as also becomes evident by the below-average proportion of religiously affiliated respondents in this group. On the whole, this group makes up no more than 15 per cent of the overall population of western Germany, however. This group with its low trust in the church and its lack of church attendance and the somewhat lower degree of trust in the church than the average Christians in western Germany seems to be little stronger in the western part of Germany as compared to the eastern part. If at the same time we take into account that churches are much stronger and religiosity is more widespread in western Germany than in eastern Germany, we can conclude that western Germany as a whole is, to a certain degree, more religious and anti-ecclesiastical than eastern Germany. This again points to the higher religious diversity of the West. If we regard the correlation between socio-structural characteristics and types of religiosity, we see clear socio-structural effects. Rather low educational levels, over-representation of rural respondents, older people, and a higher-than-average proportion of female respondents result in a higher probability that respondents will fall within groups that favour the church, more specifically the committed Christians and average Christians (cf. Table VII). Groups in which higher education levels and higher proportions of urban respondents, younger people, and men are not below average tend to be more non-religious. The comparison between non-religiousness and church affiliation, or Christian religiosity, therefore shows clear socio-structural contours, with only the church-religious group in western Germany falling out of the pattern. If individualization in the objective sense (Jagodzinski and Klein 1998) is seen as having a leveling effect on socio-structural differences, there seems to be little justification for postulating such a tendency in the religious field. The assumption of a low tendency towards individualization in the religious field is also confirmed by examining the correlation of the individualization index with socio-structural characteristics (cf. Table VII).The individualization effects found at a semantic level are clearly limited to certain social groups: groups with a below-average level of education, a stronger rural quality, higher age, and a high proportion of women exhibit particularly low individualization tendencies, while a combination of high levels of education, urban origin, below-average age levels and low numbers of women leads to the highest probability of individualization. From a socio-structural point of view, it is thus quite clear which social groups tend to favour individualization tendencies -the more modern, younger, and better-educated segments of the population. There are few committed Christians, church-religious people, or even average Christians in these segments. Conversely, there are positive correlations in British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 626 Detlef Pollack and Gert Pickel both eastern and western Germany between individualization and non-religiousness. Even religious syncretism does not profit from processes of individualization, as it seems to be less an expression of self-determination and subjective choice and more a consequence of indecision, religious indifference, and conventionality. If one assumes that processes of individualization, improved education, urbanization, and other forms of modernization have been occurring in Germany over a number of decades, one must also note that the dominant trend in the religious field has been characterized by processes of secularization - to be understood here as a loss in significance of religion in its social, institutional, and individual dimensions. Among those who are oriented towards religion, only the non-church religious group reveals a positive correlation with individualization, thus running counter to the general downward trend (cf. Table VII). We can therefore claim that the dominant process of secularization includes a reverse tendency towards non-church religious practices and ideas, which may contribute to the plural-ization and differentiation of the religious field. It should also be pointed out, however, that the non-church religious group, which is found only in western Germany and not at all in the eastern part of the country, is relatively small, and that non-church religiosity is not an alternative to traditional religiosity or Christian religiosity in either eastern or western Germany. Non-church religiosity is not a counterweight to the Christian churches and to their forms of devoutness and religiosity. It can therefore neither cushion nor compensate for the losses the Christian churches in Germany are currently suffering. Secularization is therefore - and this applies, at least for Germany, over the last 50 years - not a 'modern myth' (Luckmann 1980), even if the dominant trend described by the term secularization is slightly countered by minor reverse developments. 5. Conclusion From the 1960s onwards, Luckmann and other adherents of the individualization thesis have continuously asserted that the de-institutionalization of traditional church affiliation and religious individualization are two complementary and perhaps even compensatory processes. This assumption must be modified in various ways. Processes of de-institutionalization of religion can be easily observed by taking into account the decreasing rates in church membership, the increasing rates of church withdrawals, and the declining attendance of services (cf. Figure I and Table I). Similarly, the social relevance of individual Christian religiosity, considering belief in God or religious self-assessment as core indicators, seems to have weakened (cf. Figure II and Table II). This tendency contradicts the divergence of church adherence and individual religiosity promoted by Luckmann and his adherents. Church adherence, © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 627 measured by church membership and the frequency of church attendance, and individual religiosity, measured by belief in God and religious self-assessment, statistically correlate quite strongly. Furthermore, forms of non-church or non-Christian religiosity do not offer a serious alternative to church adherence and Christian religion. Rather, there is a close connection between non-church religiosity and individual Christian religiosity (cf. Table IV). There is, however, no significant correlation between non-church religiosity - as far as East Asian forms are concerned - and traditional church adherence. Thus, East Asian forms of non-church religiosity are not promoted by traditional church adherence, but in turn the alternative forms are also not able to replace the old ones. If we additionally take into account that only a small proportion of the overall population adheres to forms of East Asian non-church religiosity (cf. Table III), the statement that de-institutionalization and individualization of religion are complementary phenomena, which perhaps even compensate one another, becomes rather implausible. The correlation between traditional church adherence, Christian belief, and non-church religiosity is stronger in eastern Germany than western Germany. It seems that under conditions of a high degree of secularization, the various dimensions of religiosity tend to converge and to separate from the secular more strongly than in a situation of a comparatively high degree of church ties and individual religiosity in which the religious field displays a higher degree of internal plurality and is not as strictly separated from the secular. Nevertheless, tendencies that point towards a more syncretistic, non-Christian, and privatized religiosity, which gains a certain independence from traditional church adherence and religious socialization, can also be observed (cf. Table V). The relative marginality of non-church religiosity and the fact that it does not pose a serious alternative to traditional church adherence does not justify its emergence to be used as an argument against the secularization thesis. As we have seen, this kind of non-church religiosity is even interwoven with individual Christian religiosity. The trend towards secularization is predominant in Germany. It is not reversed by tendencies towards religious individualization. Rather, this tendency is a component of the ubiquitous secularization process. (Date accepted: August 2007) Notes 1. Even some proponents of the secularization thesis use the distinction between institutionalized forms of religion and individual religiosity in order to explain how secularization works. Karel Dobbelaere British Journal of Sociology 58(4) (2002: 31), for instance, delineates: Since secularization implies that 'religious institutions have lost authority and relevance in society (...) the religious situation at the individual level cannot be explained © London School of Economics and Political Science 2007 628 Detlef Pollack and Gert Pickel exclusively by secularization'. Consequently, he argues that 'the religiousness of individuals is not a valid indicator in evaluating the process of secularization'. In contrast, Berger (1969: 107f.) points out in his early work: 'The process of secularization has a subjective side as well. As there is a secularization of society and culture, so there is a secularization of consciousness.' 2. Therefore it is inappropriate to regard Luckmann as a secularization theorist, as Tschannen (1991), Swatos and Christiano (2000) or Norris and Inglehart (2004:9) have done. Only in so far as his argumentation utilizes the theoretical framework of the secularization thesis and infers from this framework the decline of the official model of religion does his approach exhibit proximity to the secularization theory. The intention of this argumentation is, however, aimed in the opposite direction. 3. This applies to the former West Germany, at least. As data is not as widely available for eastern Germany, this paper often cannot show comparable data for the eastern part of the country. Statistical data concerning the Lutheran churches in eastern Germany for the past 50 years are collected in Pollack 1994: 373ff., as far as they are available. 4. It is rather difficult to find a precise term to denote the new religious phenomena like occultism, esoterism, mysticism, New Age, energy training, Reiki, Yoga, and the like. In the literature, they are either referred to as spirituality, as new religiosity of the New Age type, as invisible religion, or even as esoteric fog (Champion 1990). 5. The survey 'Glaube und Individual-isierung (Gul)' ('Faith and individualization') was conceptualized with the aim of assessing the validity of the individualization theory. It was carried through in commission of the authors of this paper by Emnid - a leading German public opinion research institute in Bielefeld - in autumn 1999 in western and eastern Germany on the basis of a representative sample. PCE stands for 'Political Culture in Central and Eastern Europe' and denotes a representative com- parative survey carried through in autumn 2000 in 11 post-communist countries of Central and Eastern Europe, among them eastern Germany. 6. It is one of the drawbacks of Voas and Crockett's article (2005) that it does not include indicators for non-church religiousness in its analyses. 7. Respondents in eastern and western Germany tend to refute the proven correlation between traditional church affiliation and individual Christian religiosity, and partially even non-church religiosity, when they are asked directly whether they believe that religiosity can be separated from church affiliation. Asked whether it is possible to be religious without belonging to a church, an overwhelming majority of Germans in both east and west (and of Catholics and Protestants, as well as non-denominationals) give a positive reply. If, however, the same people are actually asked how many of those who do not go to church consider themselves religious, correlations become extremely weak. Only 14 per cent of those who never go to church describe themselves as religious. This is the figure for western Germany; in eastern Germany it is only 2 per cent. In contrast, 93 per cent of weekly churchgoers in western Germany regard themselves as religious, as do all eastern Germans who attend church once a week. 8. See above end of section 3, p. 11. 9. This result is in accordance with a study conducted by Dobbelaere and Voye (1996: 226), who also found a negative association between various dimensions of Catholic religiosity and expressive individualism. See also Heelas and Woodhead (2005: 111-126, esp. 115). 10. In the case of Zen meditation, it would seem that numbers are higher only because the question included both Zen meditation and Yoga. 11. The composition of the individualization index was described earlier in this paper (see the end of section 3). 12. The cluster analysis tries to differentiate between different groups of people by employing the Euclid-distances. A © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 629 correlation analysis tries to isolate similarities in the answers of separate people through comparison and depicts the strength of the correlation between variable A and variable B. In our survey, a hierarchical analysis was first used to identify the variety of clusters. Respondents were subse- quently assigned to the clusters with the par-titionary hierarchical cluster analysis. 13. The larger group of syncretists in the East (syncretists I) differs from the smaller one (syncretists II) in that they display an even higher degree of involvement in church life. Bibliography Beck, U. 1992 Risk Society: Towards a New Modernity, London: Sage. Berger, P.L. 1969 The Sacred Canopy, New York: Doubleday. Berger, P.L. 1980 Der Zwang zur Häresie: Religion in der pluralistischen Gesellschaft, Frankfurt/M.: Fischer. Bruce, S. (ed.) 1992 Religion and Modernization: Sociologists and Historians Debate the Secularization Thesis, Oxford: Oxford University Press. Bruce, S. 1999 Choice and Religion: A Critique of Rational Choice Theory, Oxford: Oxford University Press. Bruce, S. 2002 God is Dead: Secularization in the West, Oxford: Blackwell. Casanova, J. 1994 Public Religions in the Modern World, Chicago: Chicago University Press. Cipriani, R. 1989 ' "Diffused Religion" and New Values in Italy' in James A. Beckford and Thomas Luckmann (eds) The Changing Face of Religion, London: Sage. Clayton, R.R. and Gladden, J.W. 1974 The Five Dimensions of Religiosity: Toward Demythologizing a Sacred Artifact', Journal for the Scientific Study of Religion 13(2): 135-43. Champion, F. 1990 'La nebuleuse mystique-esoterique: Orientations psycho-religieuses des courants mystiques et esoteriques con-temporains' in Francoise Champion and Daniele Hervieu-Leger (eds) De l'emotion en religion, Paris: Centurion. Davie, G. 1994 Religion in Britain Since 1945: Believing Without Belonging, Oxford: Blackwell. Davie, G. 2002 Europe: The Exceptional Case: Parameters of Faith in the Modern World, London: Dartan, Longman, and Todd. Dobbelaere, K. 2002 Secularization: An Analysis at Three Levels, Brussels: Lang. Dobbelaere, K. and Voyé, L. 1996 'Europäische Katholiken und die katholische Kirche nach dem Zweiten Vatikanischen Konzil' in Franz-Xaver Kaufmann and Arnold Zingerle (eds) Vatikánům II und Modernisierung: Historische, soziologische und Perspektiven, Paderborn: Schöningh. Ebertz, M.N. 1997 Kirche im Gegenwind: Zum Umbruch der religiösen Landschaft, Freiburg; Basel; Wien: Herder. Engelhardt, K., von Loewenich, H., and Steinacker, P. (eds) 1997 Fremde Heimat Kirche: Die dritte EKD-Erhebung über Kirchenmitgliedschaft, Gütersloh: Gütersloher Verlagshaus. Evangelische Kirche in Deutschland (ed.) 1990-96 Statistische Berichte TU: Statistische Beilage Nr. 88-92 zum Amtsblatt der EKD, Hannover: EKD. Fuller, R.C. 2002 Spiritual, But Not Religious: Understanding Unchurched America, New York: Oxford University Press. Gabriel, K. 1992 Christentum zwischen Tradition und Postmoderne, Freiburg; Basel; Wien: Herder. Gabriel, K. 1993 'Wandel des Religiösen', Forschungsjournal Neue Soziale Bewegungen 6(3): 24-32. Gärtner, C. 2000 'Sinnsuche und das Phänomen der neuen religiösen Bewegung', Sociológia Internationalis 38(1): 87-113. British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 630 Detlef Pollack and Gert Pickel Gill, R. 1999 Churchgoing and Christian Ethics, Cambridge: University Press. Glock, C.Y. 1962 'On the Study of Religious Commitment', Research Supplement to Religious Education 57: 98-110. Gul 1999 Glaube und Individualisierung (Faith and individualization), Survey conducted by the Chair of Comparative Sociology of Culture at the European University Viadrina Frankfurt (Oder). Hamilton, M. 2001 The Sociology of Religion: Theoretical and Comparative Perspectives. London, New York: Routledge. Heelas, P. and Woodhead, L. 2005 The Spiritual Revolution: Why Religion is Giving Way to Spirituality, Oxford: Blackwell. Hervieu-Leger, D. 1999 Le Pelerin et le converts La Religion an mouvement, Paris: Flammarion. Hervieu-Leger, D. 2000 Religion as a Chain of Memory, Cambridge: Polity Press. Honer, A. 1985 Beschreibung einer Lebens-Welt: Zur Empirie des Bodybuilding. Zeitschrift für Soziologie 14(2): 131-9. Iannaccone, L. 1991 'The Consequences of Religious Market Structure: Adam Smith and the Economics of Religion', Rationality and Society 3(2): 156-77. Iannaccone, L. 1992 'Religious Market and the Economics of Religion', Social Compass 39(1): 123-31. Inglehart, R. and Baker, W.E. 2000 'Modernization, Cultural Change, and the Persistence of Traditional Values', American Sociological Review 65(1): 19-51. Institut für Demoskopie Allensbach 1978ff. Allensbacher Markt- u. Werbeträgeranalyse, Allensbach. Institut für kirchliche Sozialforschung des Bistums Essen (IKSE) 1999 Statistische Daten der deutschen Bistümer in Zeitreihen, Essen IKSE. ISSP 1991 and 1998 International Social Survey Programme, Köln: Zentralarchiv für Empirische Sozialforschung. Jagodzinski, W. and Klein, M. 1998 Tndividualisierungskonzepte aus individualistischer Perspektive: Ein erster Versuch, in das Dickicht der Individualisierungskon- zepte einzudringen' in Jürgen Friedrichs (eds) Die Individualisierungsthese, Opladen: Leske + Budrich. Jagodzinski, W. and Quandt, M. 1997 'Wahlverhalten und Religion im Lichte der Individualisierungsthese', Kölner Zeitschrift für Soziologie und Sozialpsychologie 49: 761-82. Jelen, T. 2002 Sacred Markets, Sacred Canopies: Essays on Religious Markets and Religious Pluralism, Lanham: Rowman-Littlefield. Kecskes, R. and Wolf, C. 1993 Christliche Religiosität: Konzepte, Indikatoren, Meßinstrumente', Kölner Zeitschrift für Soziologie und Sozialpsychologie 45(2): 270-87. Kirchenamt der Evangelischen Kirche in Deutschland 2000 Statistik kurz und bündig, Hannover: EKD. Knoblauch, H. 1991 'Die Verflüchtigung der Religion ins Religiöse: Thomas Luckmanns Unsichtbare Religion' in Thomas Luck-mann Die unsichtbare Religion, Frankfurt/ M.: Suhrkamp. Knoblauch, H. 1999 Religionssoziologie, Berlin; New York: Walter de Gruyter. Krech, V. 1998 ' "Missionarische Gemeinde": Bedingungen und Möglichkeiten aus soziologischer Sicht', Evangelische Theologie 58(3): 433-44. Krüggeier, M. and Voll, P. 1993 Strukturelle Individualisierung - ein Leitfaden durchs Labyrinth der Empirie' in Alfred Dubach and Roland Campiche (eds) lede(r) ein Sonderfall? Religion in der Schweiz, Zürich; Basel: NZN Buchverlag. Luckmann, T. 1967 The Invisible Religion: The Problem of Religion in Modern Society, New York: Macmillan. Luckmann, T. 1972 'Religion in der modernen Gesellschaft' in Jakobus Wössner (ed.) Religion im Umbruch: Soziologische Beiträge zur Situation von Religion und Kirche in der gegenwärtigen Gesellschaft, Stuttgart: Enke. Luckmann, T. 1980 'Säkularisierung - ein moderner Mythos' in Thomas Luckmann Lebenswelt und Gesellschaft, Paderborn: Schöningh. © London School of Economics and Political Science 2007 British Journal of Sociology 58(4) Religious individualization or secularization 631 Luckmann,T. 1991 Die unsichtbare Religion, Frankfurt/M.: Suhrkamp. Machalek, R. and Martin, M. 1976 Invisible Religions: Some Preliminary Evidence', Journal for the Scientific Study of Religion 15(4): 311-21. Mörth, I. 1989 'New Age - neue Religion? Theoretische Überlegungen und empirische Hinweise zur sozialen Bedeutung des Wendezeit-Syndroms' in Max Haller (eds) Kultur und Gesellschaft, Frankfurt/M.: Campus. Noelle-Neumann, E. and Piel, E. (eds) 1983 Allensbacher Jahrbuch der Demoskopie 1978-1983, Vol. VIII, München: Saur. Noelle-Neumann, E. and Köcher, R. (eds) 1993 Allensbacher Jahrbuch der Demoskopie 1984-1992, Vol. IX, München: Saur. Noelle-Neumann, E. and Köcher, R. (eds) 1997 Allensbacher Jahrbuch der Demoskopie 1993-1997, Vol. X, München: Saur. Norris, P. and Inglehart, R. 2004 Sacred and Secular: Religion and Politics Worldwide, Cambridge: Cambridge University Press. PCE 2000 Political Culture in Central and Eastern Europe, Survey conducted by the Chair of Comparative Sociology of Culture at the European University Viadrina Frankfurt (Oder). Pollack, D. 1994 Kirche in der Organisationsgesellschaft: Zum Wandel der sozialen Lage der evangelischen Kirchen in der DDR, Stuttgart: Kohlhammer. Roof, W.C. 1993 A Generation of Seekers: The Spiritual Journeys of the Baby Boom Generation, San Francisco: HarperCollins. Roof, W.C. 2001, Spiritual Marketplace: Baby-Boomers and the Remaking of American Religion, Princeton: Princeton University Press. Stark, R. and Bainbridge, R. 1985 The Future of Religion: Secularization, Revival, and Cult Formation, Berkeley: University of California Press. Stark, R. and Finke, R. 2000 Acts of Faith: Explaining the Human Side of Religion, Berkeley: University of California Press. Stark, R. and Glock, C.Y. 1968 American Piety: The Nature of Religious Commitment, Berkeley; Los Angelos: University of California Press. Stark, R. and Iannaccone, L. 1994 A Supply-Side Reinterpretation of the "Secularization" of Europe', Journal for the Scientific Study of Religion 33(3): 230-52. Swatos, W.H. and Christiano, K.J. 2000 'Secularization Theory: The Course of the Concept' in William H. Swatos and Daniel V. A. Olson (eds) The Secularization Debate, Lanham: Rowman & Littlefield. Swatos, W.H. and Olson, D.V.A. 2000 The Secularization Debate, Lanham: Rowman & Littlefield. Tschannen, O. 1991 'The Secularization Paradigm', Journal for the Scientific Study of Religion 30(4): 395-415. Tyrell, H. 1996 'Religionssoziologie', Geschichte und Gesellschaft 22: 428-57. Usarski, F. 1988 Die Stigmatisierung Neuer Spiritueller Bewegungen in der Bundesrepublik Deutschland, Köln; Wien: Böhlau. Voas, D. and Crockett, A. 2005 'Religion in Britain: Neither Believing Nor Belonging' Sociology 39(1): 11-28. Voll, P. 1993 'Vom Beten in der Mördergrube: Religion in einer Dienstleistungsgesellschaft' in Alfred Dubach and Roland Campiche (eds) Jede(r) ein Sonderfall? Religion in der Schweiz, Zürich; Basel: NZN Buchverlag. Voll, P. and Krüggeier, M. 1992 'Funktion und Substanz: Was bleibt vom freundeidgenössischen Kompromiß? Erfahrungen mit einer empirischen Untersuchung in der Schweiz', Sociológia Internationalis 30(2): 203-26. Wilson, B. 1982 Religion in Sociological Perspective, Oxford: Oxford University Press. Wohlrab-Sahr, M. and Krüggeier, M. 2000 'Strukturelle Individualisierung vs. autonome Menschen oder: Wie individualisiert ist Religion?' Zeitschrift für Soziologie 29(3): 240-4. Wuthnow, R. 1998 After Heaven: Spirituality in America Since the 1950s, Berkeley: University of California Press. British Journal of Sociology 58(4) © London School of Economics and Political Science 2007 632 Detlef Pollack and Gert Pickel Young, L.A. 1997 Rational Choice Theory and Religion: Summary and Assessment, New York: Routledge. Zinser, H. 1997 Der Markt der Religionen, München: Fink. Zulehner, P.M., Hager, I. and Polak, R. 2001 Kehrt Religion wieder? Religion im Leben der Menschen 1970-2000, Ostfildern: Schwabenverlag. © London School of Economics and Political Science 2007 British Journal of Sociology 58(4)