A history of the secularization issue

Thomas Luckmann's theory
We must now examine the separate contributions of Berger and of Luckmann to the debate on secularization. For all their agreements on many general propositions concerning the study of religion, their views on secularization differed sufficiently to lead them to apparently contradictory assessments. As will become evident, their disagreement in fact boils down to differing definitions of religion.

As the German version²⁴⁸ of the Invisible Religion was published as early as 1963, we will turn to Luckmann's theory first. In this theory, we find explicitly formulated most of the exemplars of the paradigm. Some of them - like rationalization*, pluralism, religious roots*, or collapse of the world view* - are presented cursorily. But four of them - differentiation, privatization, generalization*, and worldliness* - are presented in sufficient detail to allow us to analyze them. But before turning to this task, we must outline the general outlines of the theory.

As Luckmann almost entirely dropped his interest in religion after writing the Invisible Religion, his thought on this subject-matter did not evolve for many years, although on two occasions, he slightly rephrased his model on the occasion of conferences (Luckmann 1971; 1979). But in 1987, at the XIX^th CISR, which was entirely devoted to secularization, he introduced a new dimension. This new element constituted a new way of presenting one of the exemplars of the CISR paradigm: worldliness*. In the following pages, we will rely almost exclusively on the Invisible Religion for the presentation of the first three exemplars, but for the presentation of worldliness*, we will have to rely on the most recent papers (Luckmann 1987; 1988; 1990).

The Invisible Religion (1967) originated from Luckmann's frustration with the sociology of church-oriented religion - as practiced by Catholic sociologists (see chap. 10), or as he had practiced it himself (Luckmann 1967, p. 7). Luckmann had no quarrels with most empirical results of these studies. Indeed, the second chapter of the Invisible Religion is devoted to a summary of these findings. Briefly stated: Institutional religion has been pushed to the margins of modern society. In Europe, the persons who have most retained their religiosity are the very persons who are least relevant to the intrinsic dynamic of modern society: the young and the old, the non-working women, and the inhabitants of rural areas (1967, pp. 29-30). In the United States, religious involvement is higher, but this is obtained at the cost of an "internal secularization" (1967, p. 37).

For all their usefulness, the great mistake of these studies, Luckmann contends, lies in the equivalence put forward between church and religion. Instead of accepting the way in which the social actors define religion, we must start from a sociological - and functional²⁴⁹ - definition (1967, p. 42). The central idea underlying Luckmann's definition is that "the transcendence by the organism of its biological nature" constitutes the universal anthropological condition of religion. What does this mean?

Human beings can be considered from two points of view: as organisms (instinctual beings, incapable of detaching themselves from immediate experience), or as selves (beings capable of creating meaning in interaction with other human selves). By becoming selves, human organisms transcend their biological nature. And the transcendence of biological nature is made possible through interaction with other organisms. Admittedly, this account is fictional: In reality, human organisms do not construct systems of meaning from scratch through interaction; the system of meaning pre-exists any individual, which then gets socialized into it (1967, p. 51). But whether this transcendence of biological nature is accomplished through interaction or through socialization, Luckmann contends that this process is intrinsically religious: "It is in keeping with an elementary sense of the concept of religion to call the transcendence of biological nature by the human individual a religious phenomenon" (1967, p. 49). The "general anthropological function of religion", then, is to "mold natural organisms into persons", to "socializ[e] human beings into historical social orders" (Luckmann 1987, p. 23).

Luckmann's position in the paradigm is paradoxical, for he does not consider himself a secularization theorist. In the Invisible Religion, he referred to the "so-called process of secularization" (Luckmann 1967, p. 37). Later on, he explicitly denounced secularization theory as a "modern myth" (Luckmann [1969] 1980, pp. 161-172; 1977, pp. 17-21). Luckmann must therefore be considered a passive carrier of the paradigm. However, whenever secularization is used in a sufficiently restricted sense - as "the process which led to the increasing autonomy of different segments of the social structure, to the autonomy over against norms which had been derived from the religious cosmos" (Luckmann [1969] 1980, p. 168), Luckmann raises no objections²⁵⁰.

Luckmann's position in the paradigm stems from his functional definition of religion: Luckmann was never, from the start, interested in the fate of institutional religion. All he was interested to show was that, if properly understood and defined, religion could be considered to be intrinsically tied to man's humanity. In other words, Luckmann agrees that secularization has happened, but contends that this does not imply what is commonly taken for granted in the mythological accounts of secularization: "One may conceive of a society whose social structures perform no religious function - we live in the midst of such a type of 'secularized' society - although the fact that the social structure is 'religion-less' does not mean that a society is without religion" (Luckmann 1987, pp. 29-30).

Thus, the Invisible Religion can be read as containing two separate arguments. The first - which Luckmann finds not very interesting - is the whole account of the marginalization of institutional religion (Luckmann 1967, pp. 37-40). The second - which is the one Luckmann is personally concerned with - is the re-definition of religion in functional terms and the demonstration that religion, understood in this sense, is still very present in modern society. In the German edition, Luckmann explained: "While in the framework of the sociology of church-oriented religion, the process of secularization can be considered from the point of view of the diminishing churchliness, the general sociological attention must be directed to the 'positive' product of secularization, that is, to the symbolical relations of meaning that arise from it" (1963, p. 31). In other words, from the point of view of a believer, secularization can be seen as having two aspects: one negative, and one positive. As a matter of fact, this tension is inbuilt in the CISR paradigm. On the one hand, religion has been privatized; on the other hand, it has become generalized. Luckmann's approach very clearly articulates these two aspects.

Luckmann's account of secularization is based on the idea that the anthropological pre-condition of religion - the transcendence of biological nature - concretizes itself in a number of succeeding historical forms: 1) the elementary non-specific form (or world view); 2) the historical specific form (found in simple societies); 3) the institutionally specialized form (best represented by Christianity); 4) the institutionally non-specialized form (or "invisible religion"). The evolution leading through these different forms can be understood with reference to one of the central exemplars in the paradigm: differentiation.

Differentiation²⁵¹
Catholic sociologists tended to explain "dechristianization" by isolated factors, such as urbanization, or industrialization. Luckmann reacts against this view by subordinating these factors to the more global process of differentiation²⁵² (Luckmann indifferently uses the phrases "social differentiation", "institutional differentiation", "institutional specialization", and "institutional segmentation"). "In identifying the causes of secularization it does not suffice to refer to industrialization and urbanization as though these processes would automatically and necessarily undermine the value of traditional church religion. [...] The relation between industrialization and secularization is indirect. [...] Industrialization and urbanization were processes that reinforced the tendency of institutional specialization. Institutional specialization, in turn, tended to 'free' the norms of the various institutional areas from the influence of the originally superordinated 'religious' values" (1967, pp. 38-39). In order to avoid any misunderstanding, it should be stressed right away that Luckmann's theory of differentiation must not be read as a law of social evolution: "The differentiation of the social structure into institutionally specialized domains is not to be taken as the result of unilinear evolution. It is the product of one line of historical development which, for various reasons, took on a fateful universal significance" (Luckmann 1979, p. 128).

The initial situation* before secularization is the elementary non-specific form of religion. In this elementary state, religion is nothing but a totally undifferentiated world view, or configuration of meaning underlying a particular historical social order (Luckmann 1967, p. 51). This, however, is a hypothetical theoretical construction (1967, p. 78). In reality, in all historically known societies, religion is found in its historic specific form, which means that within the world view, a domain of meaning, which is specifically religious, has become differentiated (1967, p. 56). Thus, the rupture* causing the secularization process to unfold is situated by Luckmann so far back in time that it lies outside history.

The historic specific form of religion is found in "simple" societies (1967, p. 61). In this form, the world view is organized in different superordinated layers of meaning, of which the lowest refer to non-problematical, every-day representations (dog, tree), whereas the highest refer to "a domain of reality that is set apart from the world of everyday life" (1967, p. 61). This higher level contains in itself a representation of the structure of the world-view (as being made of superordinated layers of meaning): "Thus a 'structural' trait of the world view becomes a part of its 'contents'" (1967, p. 58). In other words, the world view now contains different levels of signification, the highest of which reflects the hierarchical organization of the world view as a whole. But in this form, there is not, as yet, any social differentiation of the religious world-view: Religion homogeneously permeates the whole social structure²⁵³ (1967, p. 61). One of the consequences of this is that one cannot, in simple societies, speak either of belief or of unbelief; the sacred cosmos is internalized as a matter of course: "There is no such thing as selective internalization - or refusal of internalization - of the sacred cosmos" (Luckmann 1971, p. 25).

The next step in the process* is marked by the appearance of the institutionally specialized form of religion, which is determined by the advent of the religious specialists. This social specialization is itself a response to the internal differentiation of the world view: "The more pronounced the distinctness of the sacred cosmos, the likelier is the emergence of a specialized institutional basis for that cosmos" (Luckmann 1967, p. 63)²⁵⁴. However, this social basis can arise only from the moment the economic system produces a sufficient surplus to allow a class of full-time religious specialists to arise (1967, p. 64). Historically, this event can be traced back to the advent of the Judaeo-Christian tradition (1967, p. 62).

But in which sense do these developments constitute steps forward in the process of secularization? In the sense, Luckmann explains, that any progress in differentiation automatically tends to undermine the plausibility of the sacred cosmos. "[Before the development of a specialized institutional basis] the 'logic' underlying the sacred cosmos is taken for granted because it is equally applicable to different social institutions. The validity of that 'logic' is reinforced by everybody. Thus the sacred cosmos and its underlying logic remain unproblematic. The chance of situations occurring in which the 'logic' of the sacred universe is no longer self-explanatory increases, however, as the social distribution of religious representations grows more heterogeneous" (1967, p. 65). The link between differentiation and secularization can thus be condensed in a formula which reads like a law: "The more unequal the distribution of religious representations, the more is the integrating function of the sacred cosmos for society as a whole threatened" (1967, p. 65). Luckmann's model is remarkably simple: everything in religion started to go awry from the moment the original world-view started to differentiate internally, thus giving rise to the necessity of a social differentiation, which in turn undermined the plausibility of the world-view.

More specifically, three factors tend to undermine the subjective plausibility of religion once it has become differentiated. The first is the appearance of tensions inside the official model: Some dimensions (for instance doctrine) pertain directly to the sacred cosmos, whereas others (for instance social ethics) pertain to the role of the church as a social institution. Thus arises the possibility of a discrepancy between them. The second factor of tension is the emergence of a specifically religious sphere in society, which results in religious practices being segregated to specific times and spaces - which carries with it the risk that individuals will consider that their religious duties can be accomplished routinely by superficial acts. Finally, the last factor is that, the official model being at the hands of full-time experts, there is a risk of their preoccupations getting increasingly remote from those of daily life. In short, "religion could be and was increasingly perceived as the ideology of an institutional subsystem" (Luckmann 1979, p. 133).

The seeds of secularization sown with the cultural and social differentiation remained dormant until the end of the middle ages (Luckmann 1967, p. 93). In this period, "Religion [...] retained a high degree of institutional specialization while the political and economic domains had not yet achieved - or regained - autonomy from the sacred cosmos"²⁵⁵ (Luckmann 1979, p. 132). How was this apparent anomaly possible? As long as social change was not too fast, Luckmann argues, there remained some possibility for the religious specialists to adapt their model to the changes that occurred. More importantly, as long as socialization to the official model took place in early childhood, individuals were not likely to question this model as they grew up (1967, p. 84). With the end of the feudal order, however, these two factors were no longer strong enough to counteract the tensions resulting from differentiation. As social change accelerated, it became increasingly difficult to adapt the official model. This in turn tended to foster the growth of secular counter-ideologies, which undermined the universal character of socialization in the official model.

Thus Luckmann's account almost entirely rests on what could be called a socio-logic of secularization, rather than on a precise historical analysis. Indeed, Luckmann's treatment of other historical forces is rather sketchy, to say the least: "Church-state relations during the period of absolute monarchies; the political and social context of the 'Religious Wars'; the proliferation of sects; the development of scientific thought and its effect on philosophic and - eventually - popular conceptions of life and the universe, the French revolution and its repercussions in 'traditionalist' and 'liberal' movements in Catholicism and Protestantism; the social consequences of the Industrial Revolution and the emergence of the working class; and the rise of ideologically oriented political parties, Biblical criticism and its effects on theology, to mention only a few factors affecting the Christian churches, attest to the danger of oversimplification" (1967, p. 94).

Privatization
The process of differentiation has three results: privatization, generalization*, and worldliness*. Let us turn, first, to privatization. Luckmann's privatization theory should not be reduced to the simple idea that the individual is left free to choose among different denominations. In Luckmann's scheme, the significance of privatization is not so much political as psychological. The whole privatization thesis rests on the central idea that institutional differentiation results in a new type of relationship between the individual and the social order (1967, p. 11). The institutional specialization of religion is only one aspect of a more global historical process in which, as the institutional order became segmented, each of these segments started to function according to specific, "rational" criteria: "The norms within the separate domains became increasingly 'rational' in relation to the functional requirements of the institution" (1967, p. 95).

Now, what are the consequences of this process for the individual? He must daily shift back and forth between different spheres, thus engaging into "a series of performances of highly anonymous specialized social roles" (1967, p. 95). As a result, the norms which he gets submitted to in these different areas are totally unrelated to his individual biography. Therefore, every individual must construct his own personal identity: "Personal identity becomes, essentially, a private phenomenon" (1967, p. 97). In the modern social order, every individual is free to determine his conduct in reference to subjective preferences, as long as he abides by the superficial rules which govern each institutional domain (1967, p. 98). Correlatively, the importance of socialization markedly decreases (1967, p. 10; 1979, p. 131), and every institutional domain "need be concerned only with the effective control of performances" (1967, p. 96)²⁵⁶.

These general considerations also apply to religion: "One highly important consequence of institutional segmentation, in general, and of institutional specialization of religion, in particular, is that the superficially religious representations, as congealed in the 'official' models of the churches, cease to be the only, and obligatory items in the sacred universe" (1967, p. 98). But even though official religion has declined, individuals continue to rely on an internalized system of "ultimate" significance: In every individual consciousness, we find a "religious" layer, which stands in the same relationship to an individual's personal identity as the specifically religious layer stands to the world view as a whole (1967, p. 71). But this system of "ultimate" significance is no longer internalized as a whole: Individual religiosity is no longer a replica of an officially given model (1967, p. 102). "Once religion is defined as a 'private affair', the individual may choose from the assortment of 'ultimate' meanings as he sees fit - guided only by the preferences that are determined by his social biography. An important consequence of this situation is that the individual constructs not only his personal identity but also his individual system of 'ultimate' significance" (1967, p. 99).
But what exactly does Luckmann mean by an "assortment of ultimate meanings"? To shed some light on this question, we must now turn to the "positive" aspect of the process of secularization: generalization*. Luckmann never uses the term generalization. I have borrowed this term from Parsons (see below) because I think it allows us to bring together a number of convergent approaches put forward by carriers of the secularization paradigm. These convergences, however, might go unnoticed because of very wide terminological divergences. The general idea is the following. The process of secularization results in two complementary movements: On the one hand, religion as it is defined by the man in the street recedes from the public into the private sphere; on the other hand, religion as it may be defined functionally by the sociologist runs out all over the world, while remaining invisible to the man in the street (and to the sociologist who stubbornly sticks to a substantive definition of religion). The first movement can be understood as a secularization in the sense that religion recedes in front of secularizing forces; the second movement can be understood as a secularization in the sense that religion permeates the world in disguise, under secular appearances. But let us come back to Luckmann's views on the subject.
Generalization*
Luckmann's views on generalization* appear under the technical label of the institutionally non-specialized form of religion - the last of the four social forms of religion. This form became more commonly referred to as the invisible religion - from title of Luckmann's book; in the book itself, however, Luckmann does not use this term. The invisible religion emerges once the institutionally specialized form has lost its preponderance. Luckmann also refers to this form of religion as the "modern sacred cosmos" or the "modern religious themes". The uncertainty in the terminology reflects the difficulty of conceptualizing this somewhat unstructured and unstable object (1967, p. 109). Luckmann recognizes that his description is somewhat tentative. The invisible religion is made up of three different components: 1) themes originating in the "private sphere"; 2) remnants of specifically religious representations; 3) remnants of former political ideologies.

The central theme originating in the private sphere is individual autonomy. This theme endows the individual's private quest for identity with an ultimate status. In other words, individual autonomy acquires a "sacred" status. Autonomy finds a number of concrete expressions, among which we find "self-realization" and "self-expression", the "mobility ethos", and a conception of sexuality as a means of defining one's own identity - rather than as an expression of the social order. Another important modern religious theme is familism - or the idea that the individual is entitled to base his private system of ultimate meanings on social interactions obtaining in the framework of the family. These themes are then surrounded by a number of other, subordinate themes, which need not be mentioned here (1967, p. 113).

The two other main components of the modern sacred universe are remnants of past formations. First, specifically religious themes are still available, even though their status has radically changed: now, they can be adopted on an individual basis, as a whole or in fragments, to enter into privately crafted systems of ultimate significance (1967, p. 105). Second, some themes which originated in the XIX^th century as economic or political ideologies are also available on the same basis (1967, p. 108).
But in which sense can it be said that these themes are "religious"? The specifically religious themes are obviously religious by definition. The former political ideologies are religious because they themselves were "late mutations of traditional religion" (1967, p. 108). Here, Luckmann simply accepts a very widespread view of generalization*. The originality of his position lies in the themes originating in the "private sphere". In keeping with his theoretical framework, Luckmann calls these themes religious because they represent one of the ways in which organisms transcend their biological nature, and become selves. By internalizing a theme like individual autonomy, the individual builds up his own system of ultimate significance, and thus becomes able to interact as a self with other selves. To consider themes born from the private sphere as religious is not strange in Luckmann's scheme; all social forms of religion are, ultimately, based on the individual: "The social forms of religion are [...] based on what is, in a certain sense, an individual religious phenomenon: the individuation of consciousness and conscience in the matrix of human intersubjectivity" (1967, p. 69).
Let us now turn to the last exemplar contained in Luckmann's theory: worldliness*. Luckmann's theory of worldliness* was already potentially contained in the Invisible Religion. As I explained in the part on differentiation, every world view is composed of superimposed layers, the lowest of which refers to unproblematic, daily perceptions (dog, tree), whereas the upper are much more abstract. In his 1987, 1988 and 1990 papers, Luckmann developed this view in a more ambitious framework, in terms of different levels of transcendence, and linked the increasing worldliness* of religion in modern society to the process of privatization.
Worldliness*
Luckmann does not use the term worldliness*, although at one point at least, he uses the expression "this-worldly" (Luckmann 1987, p. 30). His account of worldliness* is grounded in a theoretical framework based on Alfred Schutz's phenomenology. Luckmann distinguishes three levels of "transcendence": "little", "intermediate", and "great" transcendences. At each level, the transcendence consists in the fact that something which is not experienced directly by the human subject is represented indirectly - or appresented - through indicators, signs, or symbols. Thus, to give an example of a little transcendence, "the frontside of an object [...] automatically appresents its backside" (1987, p. 24). The transcendence is "little" in the sense that the thing which is appresented - the backside of the object - could itself be presented directly (for example by turning the object around). On the second level, that of the intermediate transcendence, the appresented object can no longer be perceived directly, even though it is still part of the ordinary reality of everyday life. This is the case, for example, with the experience of a fellow human being. Finally, a great transcendence is defined by the fact that the thing which is appresented "belongs to a reality different from the reality of everyday life" (1987, p. 25).

Now, these transcendences are all potentially menacing to humans: they are very close to chaos. If they want to construct meaningful existences, humans must find a way to cope with them. Humans must find a way to bridge the transcendences. A little transcendence can be bridged by indications (smoke indicates fire), an intermediate transcendence is bridged by signs, and a great transcendence can be bridged only by the use of symbols. In practice, the most common way to bridge the last two types of transcendence is through language, while rituals are generally used to bridge the great transcendences only. "Rituals place the community of interacting human beings into a relationship [...] with a kind of reality which is perceived to be extraordinary, and - among other things - extraordinarily powerful" (1987, p. 28).

Obviously, the little transcendences of everyday life are more "worldly" than the other two types, and the intermediate type is itself more worldly than the great transcendences. Although "it is the 'great' transcendences which are most obviously 'religious'", many religions are concerned primarily with intermediate transcendences, or with the transcendences of space and time. Some of the new religions (for instance the "cults" of togetherness), in particular, are concerned mainly with the construction of the intermediate, inter-subjective level of transcendence. The "historical religions", however, are concerned mainly with great transcendences.

The remainder of Luckmann's argument, which is very brief, must be quoted in full: "Conventionally institutionalized forms of religion have been replaced for large segments of the population by privatized, non-institutional meaning systems, and much of this institutionally 'invisible' new social form of religion is characterized by the fact that 'collages' of relevant religious themes depend more and more on the individual's will, and - at the same time, as the themes are linked to different levels of (socially constructed) transcendence - by the fact that stable, elaborated models integrating different levels of transcendence into a meaningful whole seem to be losing ground. In any case, the typically modern 'span' of transcendence' sems to be this-worldly" (1987, p. 30). In other words, "a wider range of different 'actors' on the social scene is now involved in the social construction of various kinds of transcendence. [...] There are the mass media, there are the churches and sects which are trying to reinsert themselves into the process of modern social construction of transcendence. [...] The tendency to shift intersubjective reconstructions and social constructions away from the 'great' other-worldly transcendences to the 'intermediate' and, more and more, also to the minimal transcendences of modern solipsism may be considered as the cultural correlate of social-structural privatization" (Luckmann 1988, p. 21; see also 1990).