CHAPTER FIVE

The Evolution of the Sociology of Religion

Theme and Variations

Grace Davie

The beginnings of the sociology of religion are barely distinguishable from the begin-
nings of sociology per se. This is hardly surprising, given that its earliest practitioners
were the founding fathers of sociology itself, all of whom were committed to the seri-
ous study of religion as a crucial variable in the understanding of human societies. Of
course, they did this from different perspectives – the outlining of which will form an
important part of the paragraphs that follow – but in the early days of the discipline,
the paramount significance of religion for human living was taken for granted, if not
universally approved. In later decades this significance was seriously questioned, not
least by sociologists of religion themselves – a fact exemplified in their prolonged pre-
occupation with the secularization thesis. In the last two decades, however, the tide
of opinion has begun to turn in a different direction, driven – very largely – by the
overwhelming (and at times somewhat frightening) presence of religion in the mod-
ern world. Given the undeniable relevance of the religious factor to the geopolitical
configurations of the new century, the sociological study of religion has gained a new
urgency. New tools of analysis and new conceptual understandings are becoming in-
creasingly necessary if sociologists are to understand (a) what is going on and (b) how
they might contribute to an evidently important debate.
This trajectory – from taken-for-granted significance, through assumed decline, to a
reestablished place in the canon – forms the theme of this chapter. It will be exemplified
in various ways, referring in turn to theoretical debate, methodological endeavor, and
substantive issues. It will, however, be overlaid, by a number of significant variations.
In the main, these relate to the different contexts in which sociologists work, contrasts
that take into account both national or regional differences and the pressures that derive
from professional obligations (research does not take place in a vacuum). It is unlikely,
for example, that a European sociologist employed by a Catholic organization in the
immediate postwar period would be preoccupied by the same questions as an American
working for a secular organization in the same decade. The fact that these two parts of
the world were, then as now, experiencing entirely different patterns of growth and/or
decline simply reinforces the point already made.
With this double aim in mind – that is, to establish and exemplify the theme,
but at the same time to take into account at least some of the major variations – this
chapter is structured as follows. It begins with an account of the founding fathers

61
62 Grace Davie

(Karl Marx, Max Weber, and Emile Durkheim), underlining their enduring legacy to the
sociology of religion – noting, however, that this legacy resonates differently. Not only
do fashions come and go, but crucially in this case, the availability of good translation is
a necessary preliminary for the great majority of readers. The lack of uniformity becomes
even more explicit as the sociology of religion moves forward: An entirely different
agenda emerges in Europe from that in the United States. The evolution in continental
(primarily Catholic) Europe concerns, very largely, the emergence of a fully fledged
sociology of religion from what has been called sociologie religieuse, a metamorphosis
that took place in a part of the world heavily influenced by decline at least in the
formal indicators of religious activity. Unsurprisingly, such debates are less relevant
in the Anglo-Saxon world, where a very different way of working has evolved. These
contrasting evolutions form the substance of the second section of the chapter.
The third will continue the contrast, introducing the two competing theoretical
paradigms in the subdiscipline: secularization theory and rational choice theory. Both
are covered in some detail in later chapters (e.g., Chapters 8 and 9). The point to be made
in this chapter concerns the emergence of two contrasting theories at different times,
in different places, to answer different questions – their roots go back centuries rather
than decades (Warner 1997). This is far from being a coincidence; sociological thinking,
like the world that it tries to explain, is contingent. The fourth and final section will
suggest, however, that the time has come to move beyond these two paradigms (with
the implication that either one or the other is correct, but not both) to more sophis-
ticated tools of analysis, if we are to understand an increasingly global phenomenon.
It is unlikely that one conceptual frame will suffice to explain all cases. A series of sub-
stantive examples will be used to illustrate both commonality and difference in the
subject matter of sociology – across a range of global regions and in a wide variety of
contexts.

THE FOUNDING FATHERS

In their sociological writing, Marx, Weber, and Durkheim were reacting to the economic
and social upheavals of the late nineteenth and early twentieth centuries, prompted
more often than not by the devastating consequences that rapid industrialization had
inflicted on the European populations of which they were part. The study of religion
could hardly be avoided within this framework, for religion was seen as an integral part
of the society that appeared to be mutating beyond recognition. Each writer, however,
tackled the subject from a different perspective (Giddens 1971; Löwith 1982; O’Toole
1984).
Karl Marx (1818–83) predates the others by at least a generation. There are two
essential elements in the Marxist perspective on religion: The first is descriptive, the
second evaluative. Marx described religion as a dependent variable; in other words, its
form and nature are dependent on social and above all economic relations, which form
the bedrock of social analysis. Nothing can be understood apart from the economic
order and the relationship of the capitalist/worker to the means of production. The
second aspect follows from this but contains an evaluative element. Religion is a form
of alienation; it is a symptom of social malformation which disguises the exploitative
relationships of capitalist society. Religion persuades people that such relationships are
natural and, therefore, acceptable. It follows that the real causes of social distress cannot
The Evolution of the Sociology of Religion 63

be tackled until the religious element in society is stripped away to reveal the injustices
of the capitalist system; everything else is a distraction.
Subsequent debates concerning Marx’s approach to religion have to be approached
with care. It has become increasingly difficult to distinguish between (a) Marx’s own
analysis of religious phenomena, (b) a subsequent school of Marxism as a form of so-
ciological thinking, and (c) what has occurred in the twentieth century in the name
of Marxism as a political ideology. The essential and enduring point to grasp from
Marx himself is that religion cannot be understood apart from the world of which it is
part; this is a crucial sociological insight and central to the evolution of the subdisci-
pline. It needs, however, to be distinguished from an overdeterministic interpretation
of Marx that postulates the dependence of religion on economic forces in mechanical
terms; this is unhelpful. The final point is more political. It may indeed be the case that
one function of religion is to mitigate the very evident hardships of this world and so
disguise them. Marx was correct to point this out. Nowhere, however, does Marx legit-
imate the destructive doctrines of those Marxist regimes that maintained that the only
way to reveal the true injustices of society was to destroy – sometimes with hideous
consequences – the religious element of society. Marx himself took a longer-term view,
claiming that religion would disappear of its own accord given the advent of the class-
less society: Quite simply, it would no longer be necessary. The inevitable confusions
between Marx, Marxism, and Marxist regimes have, however, had a profound effect on
the reception of Marx’s ideas in the twentieth century. The total, dramatic, and unfore-
seen collapse of Marxism as an effective political creed in 1989 is but the last twist in a
considerably longer tale.
In many ways, Max Weber’s (1864–1920) contribution to the sociology of religion
should be seen in this light. Rather than simply refuting Marx, Weber’s theorizing
vindicates much of what Marx himself suggested, as opposed to the vulgarizations of
later disciples. Weber stresses the multicausality of social phenomena, not least religion;
in so doing he conclusively refutes the standpoint of ‘reflective materialism’ whereby
the religious dimensions of social living simply reflect the material (Giddens 1971: 211).
But the causal sequence is not simply reversed; indeed, the emergence of what Weber
calls “elective affinities” between material and religious interests are entirely compatible
with Marx’s own understanding of ideology. The process by which such affinities come
into being must, however, be determined empirically – they vary from case to case.
Weber’s influence spread into every corner of sociology, never mind the sociology
of religion, generating a huge secondary literature – the remarks that follow are in-
evitably skeletal. Absolutely central, however, to Weber’s understanding of religion is
the conviction that this aspect of human living can be constituted as something other
than, or separate from society or “the world.” Three points follow from this (Beckford
1989: 32). First, the relationship between religion and the world is contingent and
variable; how a particular religion relates to the surrounding context will vary over
time and in different places. Second, this relationship can only be examined in its
historical and cultural specificity. Documenting the details of these relationships (of
which elective affinities are but one example) becomes, therefore, the central task of
the sociologist of religion. Third, the relationship tends to develop in a determinate
direction; a statement which indicates that the distance between the two spheres, reli-
gion and society, is being steadily eroded in modern societies. This erosion, to the point
where the religious factor ceases to be an effective force in society, lies at the heart of
64 Grace Davie

the process known as secularization – through which the world becomes progressively
“disenchanted.”
These three assumptions underpin Weber’s magnum opus in the field, The Sociology
of Religion (Weber 1922/1993), that is, his comparative study of the major world faiths
and their impact on everyday behavior in different parts of the world. Everyday behav-
ior, moreover, becomes cumulative as people adapt and change their lifestyles; hence,
the social consequences of religious decisions. It is at this point that the question of
definition begins to resonate, for it is clear that, de facto at least, Weber is working with
a substantive definition of religion, despite his celebrated unwillingness to provide a
definition as such. He is concerned with the way that the content (or substance) of a
particular religion, or more precisely a religious ethic, influences the way that people
behave. In other words, different types of belief have different outcomes. Weber goes on
to elaborate this theme: The relationship between ethic and behavior not only exists,
it is socially patterned and contextually varied. Central to Max Weber’s understanding
in this respect is, once again, the complex relationship between a set of religious beliefs
and the particular social stratum that becomes the principal carrier of such beliefs in
any given society. Not everyone has to be convinced by the content of religious teach-
ing for the influence of the associated ethic to be widespread. The sociologist’s task is
to identify the crucial social stratum at the key moment in history; it requires careful
comparative analysis.
Such questions, moreover, can be posed in ways that are pertinent to the twenty-
first century rather than the early modern period, the focus of Weber’s attentions. One
such, for instance, might engage the issue of gender rather than class or social stratum:
Why is it that women seem to be more preoccupied by religion than men at least in the
Christian West (Walter and Davie 1998)? Will the disproportionate influence of women
as the principal carriers of the religious tradition in modern Western societies have an
effect on the content of the tradition itself, or will a male view continue to dominate
despite the preponderance of women in the churches? What is the relationship between
lifestyle and belief in such societies when the roles of men and women are evolving so
rapidly?1 Such questions are just a beginning, but indirectly at least they build on the
work of Max Weber; the approach, once established, can be taken in any number of
directions. Inquiries also could be made, for example, about minority groups, especially
in societies that are both racially and religiously diverse; it is likely that minorities –
and the key carriers within them – will sustain their traditions in ways rather different
from the host society, a contrast that leads at times to painful misunderstandings.
Emile Durkheim (1858–1917), the exact contemporary of Weber, began from a
very different position. Working outward from his study of totemic religion among
Australian Aborigines, he became convinced above all of the binding qualities of re-
ligion: “Religion celebrates, and thereby reinforces, the fact that people can form
societies” (Beckford 1989: 25). In other words, his perspective is a functional one.
Durkheim is concerned above all with what religion does; it binds people together.

1
A recently published account of religion in Britain (Brown 2001) turns on precisely this point:
That is, the crucial importance of women in the religious life of Britain up to and indeed after
World War II. The 1960s and, more especially, the feminist revolution were the watershed in
this respect – no longer were women prepared to be the carriers of familial piety. Not everyone
would agree with this argument, but Brown is undoubtedly correct to highlight the significance
of gender in the analysis of religious change (and not only in Britain).
The Evolution of the Sociology of Religion 65

What then will happen when time-honored forms of society begin to mutate so fast
that traditional patterns of religion inevitably collapse? How will the essential functions
of religion be fulfilled? This was the situation confronting Durkheim in France in the
early part of the twentieth century (Lukes 1973; Pickering 1975). Durkheim responded
as follows: The religious aspects of society should be allowed to evolve alongside every-
thing else, in order that the symbols of solidarity appropriate to the developing social
order (in this case incipient industrial society) may emerge. The theoretical position
follows from this: Religion as such will always be present for it performs a necessary
function. The precise nature of that religion will, however, differ between one society
and another and between different periods of time in order to achieve an appropriate
“fit” between religion and the prevailing social order. The systemic model, so dear to
functionalists, is immediately apparent.
Of the early sociologists, Durkheim was the only one to provide his own definition
of religion. It has two elements:

A religion is a unified system of beliefs and practices relative to sacred things, that is
to say, things which are set apart and forbidden – beliefs and practices which unite
into one single moral community called a Church, all those who adhere to them.
(Durkheim 1912/1976: 47)

First there is the celebrated distinction between the sacred (the set apart) and the pro-
fane (everything else); there is an element of substantive definition at this point. The
sacred, however, possesses a functional quality not possessed by the profane; by its very
nature it has the capacity to bind, for it unites the collectivity in a set of beliefs and
practices which are focused on the sacred object. Acting collectively in a moral com-
munity, following Durkheim, is of greater sociological importance than the object of
such actions. The uncompromisingly “social” aspects of Durkheim’s thinking are both
an advantage and disadvantage. The focus is clearly distinguishable from the psycho-
logical (a good thing), but the repeated emphasis on society as a reality sui generis
brings with it the risk of a different sort of reductionism – taken to its logical conclu-
sion religion is nothing more than the symbolic expression of social experience. Such
a conclusion disturbed many of Durkheim’s contemporaries; it is still to some extent
problematic, and for sociologists as well as theologians (but see Bellah, Chapter 3, this
volume).
The evolution of the sociology of religion cannot be understood without extensive
knowledge of the founding fathers and their continuing influence (O’Toole 1984, 2000).
A further point is, however, important. The availability of their writing should not
simply be assumed; it depended (indeed it still depends) amongst other things on
competent and available translations. Willaime (1999), for example, underlines the fact
that the arrival of Weberian thinking in French sociology in the early postwar period
offered significant alternatives to those who were trying to understand the changes
in the religious life of France at this time. Weber’s work (or to be more accurate parts
of his work) became available in English almost a generation earlier (General Economic
History, The Protestant Ethic and the Spirit of Capitalism).2 It follows that a careful mapping
of the dates of translations of key texts between German, French, and English would
2
Swatos, Kivisto, and Gustafson (1998) stress an additional point. Quite apart from the ques-
tion of translation, Weber’s acceptance into English-speaking sociology was curiously delayed;
he remained relatively unknown until his discovery by Talcott Parsons. The arrival of large
66 Grace Davie

reveal interesting combinations of theoretical resources in different European societies

(as indeed in the United States). What was available to whom in the development of
theoretical thinking is not something that should be taken for granted; it could and
should be subject to empirical investigation.

THE SECOND GENERATION: OLD WORLD AND NEW

In fact, almost half a century passed before a second wave of activity took place. It came,
moreover, from a very different quarter – from within the churches themselves. Such
activity took different forms on different sides of the Atlantic. In the United States,
where religious institutions remained relatively buoyant and where religious practice
continued to grow, sociologists of religion in the early twentieth century were, very
largely, motivated by and concerned with the social gospel. A second, rather less posi-
tive, theme ran parallel; one in which religion became increasingly associated with the
social divisions of American society. The Social Sources of Denominationalism (Niebuhr
1929) and rather later Social Class in American Protestantism (Demerath 1965) are titles
that represent this trend.
By the 1950s and 1960s, however, the principal focus of American sociology lay
in the normative functionalism of Talcott Parsons, who stressed above everything the
integrative role of religion. Religion – a functional prerequisite – was central to the
complex models of social systems and social action elaborated by Parsons. In bringing
together these two elements (i.e., social systems and social action), Parsons was drawing
on both Durkheim and Weber. Or, as Lechner puts this, “Durkheim came to provide
the analytical tools for Parsons’s ambivalent struggle with Weber” (Lechner 1998: 353).
Ambivalent this struggle may have been, but Parsons’s influence was lasting; it can be
seen in subsequent generations of scholars, notably Robert Bellah and Niklas Luhmann.
The relationship with American society is also important. The functionalism of Parsons
emerged from a social order entirely different from either the turbulence that motivated
the Founding Fathers or the long-term confrontations between church and state in
the Catholic nations of Europe, most notably in France (see later); postwar America
symbolized a settled period of industrialism in which consensus appeared not only
desirable but possible. The assumption that the social order should be underpinned by
religious values was widespread.
Such optimism did not last. As the 1960s gave way to a far less confident decade, the
sociology of religion shifted once again. This time to the social construction of meaning
systems epitomized by the work of Berger and Luckmann (1966). The Parsonian model
is inverted; social order exists but it is constructed from below. So constructed, religion
offers believers crucial explanations and meanings which they use to make sense of
their lives, not least during times of personal or social crisis. Hence Berger’s (1967) idea
of religion as a form of “sacred canopy” that shields both individual and society from
“the ultimately destructive consequences of a seemingly chaotic, purposeless existence”
(Karlenzig 1998). The mood of the later 1970s, profoundly shaken by the oil crisis and its
effects on economic growth, reflects the need for meaning and purpose (no longer could

numbers of German scholars in the United States as the result of Hitler’s rise to power has-
tened a process that had already started in the 1930s. A second “renaissance” occurred in the
West as a whole in the 1980s.
The Evolution of the Sociology of Religion 67

these simply be assumed). The 1970s merge, moreover, into the modern period, a world
in which conflict – including religious conflict – rather than consensus dominates the
agenda (Beckford 1989: 8–13). Religion has not only become increasingly prominent
but also increasingly contentious.
In Western Europe, the sociology of religion was evolving along very different lines.
Religious institutions on this side of the Atlantic were far from buoyant, a situation dis-
played in the titles published in France in the early years of the war. The most celebrated
of these, La France, pays de mission (Godin and Daniel 1943), illustrates the mood of a
growing group within French Catholicism who were increasingly worried by the weak-
ening position of the Church in French society. Anxiety proved, however, a powerful
motivator. In order that the situation might be remedied, accurate information was
essential; hence, a whole series of enquiries under the direction of Gabriel Le Bras with
the intention of discovering what exactly characterized the religion of the people, or
lived religion (la religion vécue) as it became known?
Accurate information acquired, however, a momentum of its own, which led to
certain tensions. There were those, in France and elsewhere, whose work remained mo-
tivated by pastoral concern; there were others who felt that knowledge was valuable for
its own sake and resented the ties to the Catholic Church. What emerged in due course
was an independent section within the Centre National de la Recherche Scientifique,
the Groupe de Sociologie des Religions. The change in title was significant: “Religious
sociology” became “the sociology of religions” in the plural. There was, however, conti-
nuity as well as change. The initial enthusiasm for mapping, for example, which began
with Boulard and Le Bras on rural Catholicism (1947), and continued through the work
of Boulard and Rémy on urban France (1968), culminated in the magnificent Atlas de la
pratique religieuse des catholiques en France (Isambert et Terrenoire 1980). Alongside such
cartographical successes developed explanations for the geographical differences that
emerged. These explanations were primarily historical, their sources lay deep within
regional cultures. There was nothing superficial about this analysis that could, quite
clearly, be applied to religions other than Catholicism.
Willaime (1995: 37–57; 1999), Voyé and Billiet (1999), and Hervieu-Léger and
Willaime (2001) tell this primarily French (or more accurately francophone) story in
more detail: that is, the emergence of accurate and careful documentation motivated
primarily by pastoral concerns, the establishment of the Groupe de Sociologie des Re-
ligions in Paris in 1954, the gradual extension of the subject matter beyond Catholi-
cism, the development of a distinctive sociology of Protestantism, the methodological
problems encountered along the way, and, finally, the emergence of an international
organization and the “deconfessionalization” of the sociology of religion. The evolu-
tion of the Conférence internationale de sociologie religieuse, founded in Leuven in
1948, through the Conférence internationale de sociologie des religions (1981) to the
present Société internationale de sociologie des religions (1989) epitomizes this story.
It marks a shift from a group primarily motivated by religion to one that is motivated
by science, an entirely positive feature. It is, however, a story that emerges – and could
only emerge – from a particular intellectual context, Catholic Europe. Such initiatives
have been crucial to the development of the sociology of religion; they lead, however,
to preoccupations that are not always shared by scholars from other parts of the world.
The British case forms an interesting hybrid within this bifurcation: British soci-
ologists of religion draw considerably on American (English-speaking) literature, but
68 Grace Davie

operate in a European context – that is, one of low levels of religious activity. In many
ways, they face in two directions at once (Davie 2000). They are more influenced by
pluralism than most of their continental colleagues (hence a long-term preoccupation
with new religious movements rather than popular religion); this fits well with the
American literature. The parameters of religious activity in Britain are, however, very
different from those in the United States and here the work of American scholars has
proved less helpful. What is evident, however, is the inability of most (if not quite all)
British – and American – scholars to access the sociological literature in any language
other than their own. The question of translation continues to resonate. Most conti-
nental scholars can do better, leading to a noticeable imbalance in sociological writing.
Many of the latter, for example, make reference to the English-speaking literature in
their work; the reverse, however, is seldom the case until the pressure to provide an
English language edition becomes overwhelming.

THE TWO PARADIGMS: SECULARIZATION THEORY VERSUS

RATIONAL CHOICE
These differences in emphasis between European and American sociology continue into
the contemporary period, and with important theoretical consequences. Contrasting
religious situations have led not only to very different conceptual formulations but
also to a lively debate concerning the scope or range of each approach. In Europe,
for example, what has become know as the secularization thesis remains the domi-
nant paradigm (although markedly less so as time goes on); in North America, rational
choice theory has offered a convincing alternative. The substance of both these the-
ories, together with the polemics that surround them, will be considered in Part II of
this volume; there is no need to embark on that enterprise here. What is important in
terms of a chapter concerned with the different evolutions of the sociology of religion is
(a) the genesis of each theoretical outlook and (b) the scope and range of their possible
application. The two points are interrelated.
Warner’s (1993) article on a new paradigm3 for the sociological study of religion in
the United States, for example, marks a watershed in American understandings of their
own society. From this point on, the secularization thesis, already critiqued by increas-
ing numbers of scholars on both sides of the Atlantic, has to justify its applicability to
the American situation; no longer can its scope be taken for granted.4 Obviously the
process is a gradual one, and as Warner himself makes clear, his own article was part
of the process that he was trying to describe; in retrospect, however, no scholar can
afford to ignore this contribution to the literature, whether they agree with it or not.
Decisions have to be made regarding the appropriateness of secularization theory to
the American case (or indeed to any other), where once they were simply assumed.
Even more essential to a chapter concerned with sociological variations, however, is
the point introduced by Warner in the 1993 article, but considerably expanded in 1997

3
The terms “new paradigm” and “rational choice theory” are almost interchangeable. As Warner
himself makes clear, their meanings are close if not quite identical.
4
The continuing debates in the Journal for the Scientific Study of Religion, together with the col-
lection of papers brought together by Young (1997), provide ample proof of the tenacity with
which scholars, both European and American, adhere to either the secularization debate or
the new paradigm as their preferred mode of theorizing.
The Evolution of the Sociology of Religion 69

(Warner 1997: 194–6): namely, the European origins of the secularization thesis as
opposed to the American genesis of the new paradigm. The beginnings of the two
models go back centuries rather than decades. To be more precise, the secularization
thesis finds its roots in medieval Europe some eight hundred years ago. The key element
is the existence of a monopoly church with authority over the whole society; both
church and authority are kept in place by a series of formal and informal sanctions. It
is, moreover, the monopoly itself that provides the plausibility structure – the authority
is not only unquestioned, but unquestionable. Given the inseparability of monopoly
and plausibility, the latter will inevitably be undermined by increasing ideological and
cultural pluralism, a relentless process with multiple causes. Documenting this process,
or gradual undermining, is a central task of sociologists, who quite correctly describe
their subject matter (a metanarrative) as the process of secularization.
The alternative paradigm, or metanarrative, begins rather later – say, two hundred
rather than eight hundred years ago and in the new world not the old, to be more
precise in the early years of the United States as an independent nation. Here there was
no monopoly embodied in a state church, simply a quasi-public social space that no
single group could dominate. All kinds of different groups or denominations emerged
to fill this space, each of them utilizing particular religious markers as badges of identity
(religion was much more important in this respect than social class). Simply surviving
required considerable investment of time, talent, and money, not least to attract suffi-
cient others to one’s cause in face of strong competition. The possibilities of choice were
endless, and choice implies rejection as well as acceptance. The affinities with modern-
day America are immediately apparent, a situation admirably described in Ammerman’s
Congregations and Community (Ammerman 1997a). Such a book could not have been
written about Europe.
Interestingly, as Warner himself makes clear, the classics can be drawn on in both sit-
uations, although in rather different ways. Identities, for example, can be constructed
in Durkheimian terms in relation to the whole society (in Europe) or to a particular
community within this (in the United States). Likewise, Protestant sects can be seen
as undermining a European monopoly or, rather more positively, as competitors in
an American market – either way, Weber’s insights are helpful. Conversely, attempts to
impose either the secularization or the rational choice (religious economies) paradigm
wholesale on to the alternative context really do cause trouble. Such attempts arise
from a conviction that one paradigm, and only one, must be right in all circumstances.
That, in my view, is mistaken. Which is not to say that elements of each approach can-
not be used to enlighten certain aspects of the alternative situation – clearly, that can
be done and to considerable effect. A useful illustration of positive application can be
found, for instance, in Hamberg and Pettersson’s (1994) testing of the rational choice
hypothesis in different regions of Sweden. More precisely, the authors investigate the
effect of pluralism on religious activity in Sweden. Their findings support the ratio-
nal choice approach and in one of the most religiously homogeneous societies of
Europe.
The crucial point to grasp, however, lies very much deeper and illustrates, once
again, the essential difference between Europe and the United States in terms of re-
ligious understandings. More specifically, it lies in the fact that Europeans, as a con-
sequence of the state church system (an historical fact whether you like it or not)
regard their churches as public utilities rather than competing firms. That is the real
70 Grace Davie

legacy of the European past. With this in mind, it is hardly surprising that Europeans
bring to their religious organizations an entirely different repertoire of responses from
their American counterparts. Most Europeans, it is clear, look at their churches with
benign benevolence – they are useful social institutions, which the great majority in
the population are likely to need at one time or another in their lives (not least at
the time of a death). It simply does not occur to most of them that the churches will
or might cease to exist but for their active participation. It is this attitude of mind
that is both central to the understanding of European religion and extremely difficult
to eradicate. It, rather than the presence or absence of a market, accounts for a great
deal of the data on the European side of the Atlantic. It is not that the market isn’t
there (it quite obviously is in most parts of Europe, if not quite in all); it is simply
that the market doesn’t work, given the prevailing attitudes of large numbers in the
population.
What I am trying to say, using a geographical rather than sociological metaphor, is
that a map of the Rockies (i.e., more rigorous versions of rational choice theory) has to
be adapted for use in Europe – just like the map of the Alps (secularization theory) for
those who venture in the reverse direction. The map of the Rockies can, however, open
up new and pertinent questions if used judiciously and not only to test the significance
of religious pluralism strictly speaking (see Hamberg and Pettersson 1994). Interesting
possibilities emerge, for example, in the cultural as well as organizational applications
of rational choice theory (RCT) – not least with respect to televangelism. Why is it that
the European market fails to operate with respect to this particular form of religion? Or
to put the point even more directly, why has it not been possible to create a market for
this particular product? Is it simply the lack of a suitable audience or is something more
subtle at stake?5 It might, in addition, be useful to examine in more depth, and over
a longish historical period, the relationship between capital and religion in Europe.
In different historical periods, this has been extremely strong (hence, for example,
the wealth of religious art and architecture, particularly in Southern Europe – Tuscan
examples come particularly to mind). Currently, however, the relationship is weak, or
at least much weaker, although it is interesting to discover how much Europeans are
willing to invest in their religious buildings at the turn of the millennium, even among
Nordic populations where churchgoing is notoriously low (Bäckström and Bromander
1995). Used imaginatively, RCT can open up new and interesting areas of enquiry on
both sides of the Atlantic.
All too easily, however, the debate turns into a sociological fight to the death in
which one paradigm has to emerge the winner. One form of this “fight” can be found
in repeated attempts to identify the real “exceptionalism.” Is this to be the United States,
that is, a vibrant religious market in a highly developed country, but clearly without
parallel in the modern (developed) world? Or is this to be Europe, the only part of the
world in which secularization can be convincingly linked to modernization, but no
longer – as was assumed for so long – a global prototype with universal applicability?
Casanova (2001) is one author anxious to escape from this repetitious and circular
argument; we need, he argues, to think increasingly in global terms.

5
There is plenty of evidence that Europeans feared that televangelism would penetrate European
culture given the increasing deregulation of the media; in Britain, for example, it became a
major preoccupation in parliamentary debate (Quicke and Quicke 1992).
The Evolution of the Sociology of Religion 71

BEYOND THE PARADIGMS: A GLOBAL CHALLENGE

What, then, confronts the sociologist of religion who is willing to take the global chal-
lenge seriously? This question can be answered in two ways – first, by using a geo-
graphical frame, and then by considering a range of global social movements that are
essentially religious in nature. Both approaches have implications for empirical as well
as theoretical sociology and both can be found in the useful collection of essays edited
by Berger (1999).

A Geographical Perspective
In the previous sections, a firm distinction was made between the old world and the
new, contrasting both the empirical realities and the sociological thinking in Europe
with their counterparts in the United States. Without, for the time being, venturing
beyond Christianity, it is now necessary to take into account at least parts of the de-
veloping world: Latin America, sub-Saharan Africa and the Christianized parts of the
Far East (for example South Korea and the Philippines). In none of these places are the
indicators of secularization persuasive; quite the reverse, in fact, as traditional forms of
Christianity compete with innovative expressions of the faith – notably widespread and
popular Pentecostalism – for the attentions, in many cases, of growing populations. It
is true that the traditional disciplines of the Christian churches may be breaking down,
but not in favor of the secular. The movement, rather, is toward new (much less control-
lable) expressions of Christianity and emergent hybrids, notably in the Latin American
case, where an individual may be one thing in the morning (a Christian denomination)
and quite another (not least an Afro-Brazilian variant) in the afternoon. Add to this al-
ready extensive list the parts of the world dominated by other world faiths – the hugely
varied Islamic nations, the competing religious traditions of the Middle East, the Sikhs
and Hindus of the Indian subcontinent and the great diversity of Eastern religions –
and Berger’s claim that the developing world is “as furiously religious as ever” seems
well justified (Berger 1992: 32).
In geographical terms, the only possible exceptions to a religious worldview are
Japan and West Europe, together with West Europe’s outposts in the form of the English-
speaking Dominions – all of which, it is important to note, constitute developed global
regions. (The great unknown remains, of course, the immense Chinese population, in
which it is still difficult to predict what is likely to happen in religious terms both in the
short and long term.) The fact that the two most secularized parts of the globe are two
of the most developed does, however, give pause for thought regarding the possible
connections of modernization and secularization – the core of both modernization
and secularization theory (Inglehart 1990, 1997).6 These cases, however, need to be
balanced against the United States, which – it is abundantly clear – remains a very
notable exception; the relationship is by no means proven.
The situation is, in fact, confused rather than clear-cut, a fact revealed in the rich
selection of material brought together in Heelas and Woodhead (2000), and increasingly
in the most recent textbooks concerned with the sociology of religion (Aldridge 2000).

6
Interestingly, Inglehart’s most recent account is rather more nuanced. Economic moderniza-
tion is indeed associated with value change, but such change is path dependent. In other words,
the broad cultural heritage of a society (not least the religious element) leaves an imprint that
endures despite modernization (Inglehart and Baker 2000).
72 Grace Davie

It becomes increasingly apparent, for example, that different trends may well coexist
within the same society, quite apart from the contrasts between different global regions.
We need tools of analysis that are able to cope with this complexity.

Thematic Approaches
A thematic approach to the same question tackles the material from a different per-
spective – looking in turn at three global social movements: (a) global Catholicism,
(b) popular Pentecostalism, and (c) the possibly overlapping category of fundamental-
ism (encompassing a variety of world faiths).
Casanova (2001) points out the paradox in modern currents of Catholicism. At pre-
cisely the moment when European expressions of Catholicism begin to retreat almost
to the point of no return – as the convergence between state and church through cen-
turies of European history becomes increasingly difficult to sustain – Catholicism takes
on new and global dimensions. It becomes a transnational religious movement, and as
such has grown steadily since 1870 (the low point of the European Church). The Papal
Encyclicals from this time on are concerned primarily with the dignity of the human
person and with human (not only Catholic) rights, a movement that accelerates rapidly
as a result of the Second Vatican Council. Transnational Catholic movements begin to
grow (for example, Liberation Theology, the Opus Dei and Communione e Liberazione),
centers of learning become equally international, so, too, does the Roman Curia emerg-
ing as it does from cross-cutting, transnational networks. One aspect of such links is the
growing tendency toward movement, manifested among other things in the increasing
popularity of pilgrimage. Most visible of all, however, is the person of the Pope himself,
without doubt a figure of global media proportions. The Pope goes nowhere without
planeloads of the world’s media accompanying him, and his health is the subject of
constant and minute speculation in the international press. Conversely the capacity of
the Pope to draw huge crowds of Catholics (not least young people) to one place can
be illustrated in the World Youth Days that took place as part of the millennium cele-
brations in Rome 2000: Two million young people came together in the final all-night
vigil and Sunday morning mass at the Tor Vergata University (August 19–20). Few, if
any, secular organizations could compete with these numbers.
It is hardly surprising that the different elements that make up this increasingly
global movement attract negative as well as positive comments. That is not the point.
The point is the existence of a transnational form of religion with, at the very least,
considerable influence on a wide range of moral and ethical debates, crucial factors for
the sociologist of religion at the beginning of the twenty-first century.
Global Pentecostalism is rather different in that its immediate impact is less visible.
Its effect on huge and probably growing numbers of individuals is, however, undeni-
able, a phenomenon that is attracting the attention of increasing numbers of scholars
and in a variety of disciplines. The literature, as a result, is growing fast (see, for exam-
ple, Corten 1997).
Coleman (2001), Freston (2001), and Martin (2002) offer state-of-the-art accounts of
this phenomenon, each concentrating on a different dimension. Coleman, for example,
is primarily concerned with “Health and Wealth” Christians and how they establish
effective global communications, not least by means of electronic technologies. Freston
concentrates on the political dimensions of evangelical Christianity, an aspect that is
particularly difficult to discern given the fragmented, fissiparous, and often apolitical (at
The Evolution of the Sociology of Religion 73

least in a conventional sense) nature of the movement. Martin, in contrast, is concerned

first and foremost with the cultural aspects of Pentecostalism, and more especially with
cultural change. His book is wide ranging, covering the diaspora populations of the
Far East in addition to North America, Latin America, and sub-Saharan Africa. The
movement of Pentecostal Christians from one part of the world to another and the ways
in which their churches enable such migrations (both culturally in terms of motive and
organizationally in terms of welcome) provides an important cross-cutting theme.
As a postscript to this discussion, it is important to note that, in a developed theo-
retical chapter, Martin pays considerable attention to the absence of Pentecostalism as
a widespread and popular movement in both Europe and the United States. Currents
of Pentecostalism do, of course, exist in Europe – both within and outside the histori-
cal churches – but they are not large in numerical terms (nor in consequence all that
influential). In terms of global Pentecostalism, the notion of European exceptionalism
appears to gain a certain credibility.7 The American case is rather different. Here it is
the vigorous nature of the evangelical constituency that is resistant to newer forms of
Pentecostalism. In other words, the movement exists, but is substantially contained
within the existing denominations; there is no need to “walk out” (Martin’s term) into
new forms of religious organization to find salvation. It follows, however, that neither
the European nor the American experience will be all that helpful in understanding
popular Pentecostalism in other parts of the world; there is a need for more innovative
sociological thinking.
Fundamentalism(s) – whether in the singular or in the plural – is one of the most
controversial and debated terms in both academic and popular discussion. One focus
of this debate concerns the largely unresolved issue of whether a term that was used
initially to describe currents of conservative Protestantism popular in the early twen-
tieth century in parts of the United States can be helpfully transposed to a series of
trends visible in a variety of world faiths some sixty or seventy years later (a theme
that picks up the central argument of this chapter). The fact that the terminology is
difficult should not, however, detract from the evidence that these trends are indeed
taking place – reversing in many ways the expectations of the Western (often European)
observer, who assumed not only decreasing levels of global religiosity as the twentieth
century drew to a close, but that such religion as continued to exist would manifest
increasingly “reasonable” tendencies.
That did not happen, at least not universally. What has happened – in different
places and in different world faiths – has been the emergence of a range of reactive,
conservative religious movements, resisting, in some cases, the modernizing trends
evident within the major faiths (modern biblical criticism, for example) or, in others,
the incursions of modernization (very often associated with secularization) from the
outside. Once again, the scholarly literature is immense. A huge, and – to some extent –
representative set of volumes (although not everyone would agree with its findings)
can be found in the Fundamentalism Project, published through the early 1990s by the
University of Chicago Press (Marty and Appleby 1995). For our purposes, two aspects
can be drawn from this vast accumulation of scholarship: First, the discussion of the

7
Partial exceptions to the exception exist on the margins of Europe. See, for example, the gypsy
population of parts of central Europe and the interesting case study taken from Southern Italy
quoted in Martin (2002).
74 Grace Davie

concept of fundamentalism itself and, second, the range and location of the case studies
which form the heart of the empirical project.
The great variety of movements that are considered under the heading of “fun-
damentalism” display what the authors call “family resemblances” – leading to the
creation of an “ideal-type” (in the Weberian sense) of fundamentalism, against which
any particular case can be measured. Not all examples will meet all the criteria set out,
but in order to be included they need to meet a minimum number. Several subtypes
emerge within the overall concept. The important point to grasp, however, is the no-
tion of fundamentalism as a “heuristic device,” which enables us to examine – not, it
is important to remember, always to approve – a wide variety of religious movements
currently active in the modern world.
The associated case studies are taken from all the major world faiths and from almost
all parts of the globe. What, however, is striking from the point of view of a chapter
concerned with the different contexts of sociological development is, once again, the
relative absence of examples from Europe, although not in this case from America. The
three potential candidates for Europe are the following: Traditional “Lefebvre type”
Catholicism, Ulster Protestantism, and the Italian-based youth movement – Commu-
nione e Liberazione (already mentioned in connection with global Catholicism). In terms
of the ideal-type of fundamentalism, however, none of the three fit the criteria com-
pletely or convincingly. The first is closer to traditionalism than a reactive fundamen-
talist movement, the second is more of an ethnic nationalism than a social movement,
and the last has been described by Italian commentators (Pace and Guolo 1998) as a
“fondamentalismo ben temperato”; it is, in other words, a partial illustration of fun-
damentalism, displaying some of the “family resemblances” but lacking, in particular,
any sustained reference to a sacred text.
The American case raises rather different issues, some of which connect very directly
with the rational choice paradigm introduced in the previous section. Conservative
forms of Protestantism (and perhaps of Catholicism, too), whether these are full-fledged
fundamentalisms or not, quite clearly form an important part of the American religious
market. One of the most successful applications of rational choice theory, moreover,
has been to explain the relative popularity of conservative (high cost) as opposed to
liberal (low cost) choices in the religious life of the United States (Iannaccone 1992a,
1994). Rather more problematic, however, are the attempts to apply the same type of
theorizing to forms of fundamentalism found outside America – in places, for example,
where the concept of the market has virtually no resonance. In such cases, additional
factors have to be taken into account to understand the reactive, high cost, and, at
times, violent nature of religious activity.
Bearing such complexities in mind, how should the sociologist of religion working
at the beginning of the new century proceed? First, surely, by acknowledging the ur-
gency of the task – we need to understand what is happening given the salience of the
religious factor in geopolitics of the modern world. Then, perhaps, by returning to the
essentially Weberian statement invoked in the introductory paragraphs of this chapter,
namely that sociological thinking, like the world it tries to explain, is contingent. With
this in mind, it is very unlikely that one theoretical frame – be it European or American
or another – will fit all cases. This does not mean that either secularization theory
or rational choice theory should be totally abandoned; they should however be used
The Evolution of the Sociology of Religion 75

judiciously and are likely to make more sense in some parts of the world than in
others.8
Whatever the theoretical difficulties, it is abundantly clear – given the nature of
the religious phenomena described in this section – that a global frame of reference
is increasingly necessary. And within such a framework, careful comparative analy-
sis becomes the most obvious way to work if we are to reveal the specific features of
particular cases, from which accumulations of data begin to emerge. Patterns and con-
nections begin to form, which in turn suggest heuristic (and sometimes full-fledged
theoretical) possibilities, for example the “ideal type” of fundamentalisms already out-
lined. Martin’s work on global Pentecostalisms offers another example (Martin 2002).
Building from encyclopedic reading in the field, largely of relatively small-scale an-
thropological studies, Martin constructs a framework through which to “make sense”
of these very different situations. The framework is strong enough to guide the reader’s
thinking, but sufficiently flexible to allow the empirical material to speak for itself.
Among many emergent themes, Martin makes it abundantly clear that circumstances
alter cases, once again underlining the essential point: The world is indeed contingent
and effective sociological thinking must take account of this fact, if it is to understand
(or even begin to understand) the bewildering variety of ways in which religion and
modernity interconnect.

8
It is interesting, for example, that Finke and Stark’s contribution to this volume makes reference
to both the Latin American (Pentecostal) and the fundamentalist cases. RCT undoubtedly sheds
light on these examples (especially the former), alongside other theoretical perspectives. Gill’s
work (1998, 1999) on Latin America is particularly helpful in this respect.