Some Second Thoughts on Substantive versus Functional Definitions of Religion

Author(s): Peter L. Berger

Source: Journal for the Scientific Study of Religion, Vol. 13, No. 2 (Jun., 1974), pp. 125-
133
Published by: Wiley on behalf of Society for the Scientific Study of Religion
Stable URL: http://www.jstor.org/stable/1384374
Accessed: 19-05-2016 10:42 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].

Wiley, Society for the Scientific Study of Religion are collaborating with JSTOR to digitize,
preserve and extend access to Journal for the Scientific Study of Religion

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
 Some Second Thoughts on
Substantive versus Functional
Definitions of Religion*
PETER L. BERGER

Department of Sociology
Douglass College, Rutgers University

Scientific approaches to religion have always alternated between functional and substantive
definitions of the field-that is, between defining religion in terms of its social or psychological
functions and in terms of its believed contents. Recently there has been a predominance of
functional definitions, as exemplified by Bellah, Geertz, and Luckmann. Quite apart from the
scientific utility of these definitions, they have come to serve an ideological use-as a quasiscientific
legitimation of the avoidance of transcendence. This is in accord with a secularized Zeitgeist, but it
threatens to lose sight of the very phenomenon of religion. To regain the phenomenon, what is
required is a return to a substantive definition, an understanding of religion "from within."
Schutz's analysis of "multiple realities" in human experience may serve as a useful starting point for
this.

M y purpose here is to state some second thoughts on a topic about which I have
written in the past, namely, the definition of religion within the framework of its
scientific study. I do so with just a little reluctance, because in the last few years my
work has been taking me elsewhere, to the sociology of development in Third World
countries-an absorbing and deeply troubling area, leaving rather little mental energy
for other topics. Nevertheless, I have found that the concern to understand religion is a
chronic rather than acute condition-at best, it remains dormant for a while.
First, I would like to state some convictions about which I have never changed my
mind, and especially one conviction: The scientific study of religion must bracket the
ultimate truth claims implied by its subject. This is so regardless of one's particular
conceptions as to scientific methodology-for instance, as between "positivistic" or
"humanistic" conceptions of science. If science means anything at all, as distinguished
from other types of mental activity, it means the application of logical canons of
verification to empirically available phenomena. And whatever else they may be or not
be, the gods are not empirically available, and neither their nature nor their existence
can be verified through the very limited procedures given to the scientist. What is
available to him is a complex of human experience and thought that purports to refer
to the gods. Put differently, within the framework of science the gods will always

*This is a very slightly revised version of an address delivered at the 1973 meetings of the American Academy
of Religion, Chicago.

Copyright ? 1974 by the Society for the Scientific Study of Religion

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
 126 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

appear in quotation marks, and nothing done within this framework permits the
removal of these quotation marks. Anyone engaged in the scientific study of religion
will have to resign himself to this intrinsic limitation-regardless of whether, in his
extrascientific existence, he is a believer, an atheist or a skeptic.
This means that, within the framework of science, transcendence must appear as
immanence. The gods, which appear in the religious consciousness as possessing an
ontological status transcending this consciousness, are not available to the scientist in
this alleged status-they are only available qua contents of human consciousness, thus
as immanent by necessity. If one likes the language of Feuerbach and Marx, then, it is
quite proper to say that, in the framework of science, religious meanings must appear
as human projections. The philosophical question will then be whether they may also
be something else (and here, I think, neither Feuerbach nor Marx are very helpful). The
scientific question, however, will be the understanding and location of these meanings
within human experience. And here the scientist's methodology becomes important.
My own methodology, in this as in any other area of human affairs, is based on the
fundamental "humanistic" proposition that the human world is essentially a network
of meanings and that, therefore, nothing in this world can be adequately understood
without understanding of these meanings "from within". This, of course, is what Max
Weber called Verstehen. In the case of religion, to be sure, the transcendent meanings
will be "translated" into terms of immanence. The question is whether this
"translation" takes into account the intentions of those who adhere to these meanings,
or whether it distorts them in the process of transposition. This question begins to be
answered, in different ways, as soon as a definition of the religious phenomenon is
posited.
In the 19th century it was common to have generic definitions of religion which
implied a reductionist philosophy of religion from the begining. The method might be
called assassination through definition. Thus Max Mueller characterized religion as a
"disease of language" (in this, incidentally, preempting in an interesting way some
20th-century approaches of linguistic philosophy), Edward Tylor as some sort of
imperfect philosophy, and Marx and his followers as a root manifestation of "false
consciousness." These approaches to definition have been rightly discredited in the
more recent scientific study of religion (interestingly enough, even by some Marxists
and linguistic philosophers). Instead, in recent decades, there have been two
predominant approaches to the definition of religion. Religion has been substantively
defined, in terms of the meaning contents of the phenomenon. And it has been
functionally defined, in terms of its place in the social and/or psychological system.
The substantive approach, which has always been at least tacitly assumed by most
historians of religion, was greatly refined by the phenomenological school of
Religionswissenschaft. Rudolf Otto, Gerardus van der Leeuw and Mircea Eliade
should be specially mentioned. The functional approach was, as it were, codified in the
social sciences by Emile Durkheim and Bronislaw Malinowski, and was given
enormous impetus in the psychology of religion by the influence of Freud and his
various disciples. In recent American scholarship, especially in the social sciences,
there has been a strong preference for the functional approach.
The definitions of religion employed by Robert Bellah (1964), Clifford Geertz
(1966) and Thomas Luckmann (1967) represent this functional approach in different

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
SECOND THOUGHTS ON DEFINING RELIGION 127

versions. Thus Bellah (clearly influenced by Paul Tillich in this) defines religion as
"A set of symbolic forms and acts which relate man to the ultimate condition of his
existence" (1964: 358). Geertz has a more complex definition of religion as "a system of
symbols which acts to establish powerful, pervasive, and long-lasting moods and
motivations in men by formulating conceptions of a general order of existence and
clothing these conceptions with such an aura of factuality that the moods and
motivations seem uniquely realistic" (1966: 4). Both Bellah and Geertz reflect the
Durkheimian impetus in the sociology of religion, as refracted through British social
anthropology and American structural-functionalism in sociology (especially, of
course, as mediated through the work of Talcott Parsons, whose students both of them
were at Harvard). Luckmann's version of the functional approach is different,
reflecting a very interesting combination of Durkheim (without the Parsonian
variations on his basic themes) with recent European philosophical anthropology. The
key phrase in Luckmann's version is the characterization of religion as "the
transcendence of biological nature" (1967: 49) (not to be confused with transcendence
in the conventional religious sense). The influence of Helmuth Plessner is particularly
important for Luckmann's approach to religion.
Both the Bellah/ Geertz and the Luckmann versions of definitional functionalism
are highly sophisticated conceptual constructions, and it is impossible to do justice to
them here. I only want to focus on what is essential for my present purposes: In all three
cases, religion is defined in terms of what it does-be it for society, for the individual, or
for both. And this, of course, is what the word "functional" essentially means.
Now, I have never liked functional definitions of religion of this kind. For one thing,
they are very broad, and I've always been convinced that useful definitions are narrow.
More importantly, functional definitions have at least the tendency to violate the
methodological premise I learned from Weber- that any human meaning must, first
of all, be understood in its own terms, "from within," in the sense of those who adhere
to it. But my old position (I stated it briefly in an appendix to my book, The Sacred
Canopy in 1967) was that there is little point to arguing about definitions. After all,
definitions are always ad hoc constructs. They don't fall from heaven. They have a
specific cognitive purpose. To some extent, definitions are a matter of taste.
Consequently, my attitude to different definitions of religion was one of relaxed
ecumenical tolerance. It is this attitude that I would like to revise now: I have become
more militant in my opposition to functional definitions. This is not due to a change in
my conception of science and its methodological presuppositions. Nor is it due to any
lessening in my esteem for the work of many who use functional definitions, and
especially for the work of Luckmann, Geertz and Bellah. I have learned a lot from all
three and (apart from the matter at hand) find myself in agreement with many of their
insights into religious phenomena in the contemporary world. The reason for my
revised attitude is rather that I have become more aware of the ideological uses of
functional definitions of religion. And I would like to stress very strongly that by this I
don't mean the personal worldviews of Luckmann, Geertz or Bellah, but rather the
uses to which their approaches have been put by others in the marketplace of ideas.
Definitions focus intellectual attention. Definitions "slice up" reality in different
ways. As between functional and substantive definitions of religion, this becomes very
clear if one looks for what they, respectively, include and exclude. Thus substantive

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
128 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

definitions of religion generally include only such meanings and meaning-complexes as

refer to transcendent entities in the conventional sense-God, gods, supernatural
beings and worlds, or such metaempirical entities as, say, the ma'at of the ancient
Egyptians or the Hindu law of karma. By contrast, functional definitions are likely to
include such meaning-complexes as nationalism, or revolutionary faiths, or the mobil-
ity ethos, or any number of new "life-styles" with their appropriate cognitive and
normative legitimations. One result of this casting of a very broad definitional net is
quite simple. The commonalities between, say, the Hindu metaphysic and the new
"sensitivity" of the contemporary counterculture are brought into focus-and, by the
same token, the differences between them are relegated to inattention.
As long as such questions are negotiated within a strict methodological context (in
which possible ideological uses of scientific conceptualizations are irrelevant), I would
continue to say that utilitarian criteria will govern the choice of definition (utilitarian,
of course, in a theoretical rather than practical sense). For one theoretical purpose, a
commonality between empirical data may be more important than the difference
between them-and the reverse may hold for another theoretical purpose or
intellectual enterprise. Thus I continue to concede the utility of functional definitions
for certain types of inquiry-for example, investigations of the social-psychological
mechanisms by which this or that worldview is maintained as plausible in the minds of
its adherents. In such an investigation, it makes a lot of sense to compare the
plausibility-maintenance rituals of, say, adherents of a traditional religious faith, of a
political doctrine, or of a deviant style of sexuality. For another example,
investigations of the manner in which comprehensive world views legitimate social
inequities may well wish to compare, say, traditional religion and various political
creeds without paying much attention to their differences. In other words, in such
instances the scientific investigator may validly refrain from certain refinements of
Verstehen, and from the differentiations to which the latter leads.
But what has intrigued me of late is something else: To wit, there seems to be a
positive motive to refrain from Verstehen in the area of religion! And this, I would
contend, has reasons over and beyond theoretical utilitarianism, reasons grounded in
vested interests of an ideological character.
There is one rather innocent ideological interest that can be passed over quickly. It
comes from what may be called the imperialism of academic disciplines, and it is often
relevant to the scholarly projects of individuals. Thus there are sociologists who know
very little about religion but who want to study a particular phenomenon with religious
overtones. Conversely, there are scholars with religion degrees who want to study, say,
communes, or changing sexual mores, or political movements. Both sets of individuals
have a mild ideological interest in subsuming under the same definition their earlier
field of specialization and the area into which they are now branching out. This is not
terribly interesting. Let me only say that I am very much in favor of some people
branching out, but they can do this armed with just about any conceptual apparatus,
and their real problem is not in the realm of intellect but on the job market.
The ideological interest that concerns me is much more basic: It is the interest in
quasiscientific legitimation of the avoidance of transcendence. My thesis is this: The
functional approach to religion, whatever the original theoretical intentions of its
authors, serves to provide quasiscientific legitimations of a secularized world view. It

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
SECOND THOUGHTS ON DEFINING RELIGION 129

achieves this purpose by an essentially simple cognitive procedure: The specificity of

the religious phenomenon is avoided by equating it with other phenomena. The
religious phenomenon is "flattened out." Finally, it is no longer perceived. Religion is
absorbed into a night in which all cats are grey. The greyness is the secularized view of
reality in which any manifestations of transcendence are, strictly speaking, meaning-
less, and therefore can only be dealt with in terms of social or psychological functions
that can be understood without reference to transcendence.
Against this tendency to liquidate the religious phenomenon by defining it out of all
specificity, I would recommend that the scientific study of religion return to a
perspective on the phenomenon "from within," that is, to viewing it in terms of the
meanings intended by the religious consciousness. I rather doubt that, as far as the
definitional delineation of religion is concerned, it will be possible to go very far
beyond the contributions of the phenomenological school. Indeed, I think that one
could do worse than to return to Otto's starting point in this matter. There is one
important theoretical angle, however, that remains to be explored: This is the
delineation of religious experience as against other realms of experience, and most
importantly as against the experience of ordinary, everyday reality. In such an
undertaking, I believe, phenomenology will again be particularly useful, especially
Alfred Schutz's analysis of the "multiple realities" of human experience. This would be
a rather formidable theoretical task, and it cannot even be begun here, but it might be
useful to make a few suggestions as to the procedure this would require.
One would begin with an understanding of the ordinary, everyday "life-world," the
taken-for-granted world of commonsense, which Schutz aptly called the "paramount
reality." This is the reality that we apprehend when we are "wide awake," when we go
about our various mundane projects (Schutz also called this the "world of working"),
and which we presuppose as the common context of almost all social interaction. This
is the reality that we massively share with our fellowmen, and which therefore is the
most massively plausible to ourselves: Almost everyone around us, almost all the time,
confirms and reconfirms the basic contours of this reality. It is thus a domain of
familiarity and of safety-indeed, it is the principal domain of reliability in our
experience. For this reason, we have a very strong interest in maintaining its
plausibility. There is hard psychological evidence to the effect that any perceived
threats to the taken-for-granted reality of everyday life are met with acute anxiety and
with the prompt erection of cognitive defenses. Without the plausibility of this
"paramount reality" there could be no collective life, no society, and in all likelihood
even individual existence would be unbearable without it for most people.
This ordinary, safe reality of everyday experience "contains" other realities. Schutz
called these "finite provinces of meaning." From the standpoint of everyday life (which
is the standpoint we occupy most of the time) these other realities appear as enclaves,
inslands, "holes." As such, they always constitute an implicit threat to the taken-for-
granted security of the ordinary. It is, therefore, very important to hold these other
realities within "boundaries," which are culturally defined and institutionally
organized. The realities of dreams, of pure intellectual activity, and of aesthetic ex-
perience are cases in point. If mundane activity is to go on (from agriculture to poli-
tics), the borders between the reality of dreams and the reality in which we are "wide
awake" must be maintained (which, by the way, is no more true today than it must have

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
130 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

been in the Stone Age). If the business of society is to go on, closure must be imposed
on the more searching intellectual questions (philosphers have the propensity to
filibuster and, in sheer self-defense, it is easy to see why the Athenians had to get rid of
Socrates). If we are not, literally, to go out of our minds individually and collectively,
the more intense aesthetic ecstasies must be properly scheduled and administered in
manageable dosages (Plato was perfectly right in his suspicions about music). The
"migrations" of individual consciousness between these spheres of reality cannot be
prevented: Even if all intellectual and aesthetic activity were successfully proscribed,
we would still sleep-and, sleeping, we would dream. However, as long as the
boundary defenses are in reasonably good working order, the individual will always
"return to reality" from his wanderings into other worlds-he will return, that is, to the
world of everyday life that is socially constructed as the ens realissimum.
The reality of everyday life is ever again breached, as other realities force
themselves upon consciousness. The ordinary world, including the individual's own
identity in it, then becomes "unreal," at least temporarily: In this moment, a nightmare,
or the universe of theoretical physics, or the atmosphere of a particular novel, are
"more real" than the world in which I hold a job as a professor, in which I worry about
my toothache, and in which I have parked my automobile; conversely, these mundane
concerns lose their "accent of reality" (to use William James' suggestive phrase). This
side of what we call psychosis, at any rate, human experience is an ongoing succession
of resting securely within the reality of the ordinary, having that reality breached, and
returning to the ordinary after the breach in its defenses has been repaired. Different
societies employ different institutional means to exorcize the demons that lurk in every
such breach; no human society can dispense with such exorcism altogether.
It will be clear from the preceding examples that by no means is every breach in the
fabric of everyday reality religious-nor are all the "extra-ordinary" realities into
which the individual is, from time to time, transposed. Still I strongly suspect that most
if not all breaches of the "paramount reality" are potential mediators of religious
experience. As Otto has conclusively shown, the religious is the realm of "otherness,"
of the "uncanny": Every breach in what is ordinarily taken for granted as "canny" at
least potentially propels consciousness toward the "other." Let me use a picture here:
Ordinary life is sitting and moving within a well-lit house, with familiar fellowmen, in
an atmosphere of warm security. Religious experience takes place in the night outside.
Every breach in the walls of the house implies at least the possibility of this particular
nighttime encounter. (Let me say in passing that this picture by no means excludes the
many cases of, so to speak, ordinary houses with religious ornaments, even with a
religious atmosphere. But these cases already constitute "containment": The
"otherness" of the religious experience is, literally, "domesticated.")
The specificity of religious experience, then, is not to be sought in its breaching of
the "paramount reality," but rather in the characteristics that structure its "finite
province of meaning." To repeat, I doubt whether much progress can be made beyond
Otto in delineating these characteristics. However, in addition to Otto's criteria of the
"numinous," the relation of the latter to the "paramount reality" reveals some further
characteristics. The relation may be described as follows: In the context of religious
experience, the reality of everyday life is dramatically deprived of its "paramount"
status. Instead, it now appears as the antechamber, the "outer court," of another

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
SECOND THOUGHTS ON DEFINING RELIGION' 131

reality, one of drastic otherness and nevertheless of immense significance for man. In
the course of this shift in the apperception of reality, all mundane activity in everyday
life is radically relativized, trivialized-in the words of Ecclesiastes, reduced to
"vanity." Despite this relativization, the individual must continue to act in everyday
life, and to this end must somehow continue to behave "as if" its reality still had its old
status. Put simply: Even a saint must eat. This paradox of being in two realities more or
less simultaneously is the fundamental problem of the religious life. There are, of
course, very different attempts at solution in the history of religion.
There are great variations in intensity in this experience. It is not at all my intention
to say that only the most intense versions (such as the events reported by the great
mystics) should be designated as genuine religious experiences. The crucial shift in the
"accent of reality" even occurs in experiences of moderate emotional texture-for
instance, in the faithful performance, yet another time, of a religious ritual. If such a
ritual is accompanied by the subjective intentionality appropriate to it, the
relativization of everyday reality is ipsofacto posited. Apart from ritual, there is a great
variety of intutions, intimations, approximations of the basic reordering of realities.
And even when the other reality is not "at hand" to consciousness, it is copresent with
the reality of everyday life-lurking in the background, so to speak. What this means is
very well expressed in a German word that derives from the theater:
Doppelboedigkeit-literally, the quality of having two levels. Religion introduces an
additional dimension to experience; once this has happened, there continues to be the
awareness that the stage of everyday action is "hollow," that there is another level
"beneath" it, and that the figures located there may "surface" at any moment. Needless
to say, this awareness (even if at times it is only held dimly) accentuates the
precariousness of evervdav life and all its works.
Everything we know historically indicates that this experience is one of the oldest,
perhaps as old as the human race. It is what Robert Musil, in his novel The Man
Without Qualities (1965), called "the other condition" (der andere Zustand), coeval
with the condition of being in the world of ordinary reality. (Allow me to remark in
passing that Musil's work, in addition to its many other great qualities, may be
regarded as the most important religious novel of this century.) Musil has ingeniously
described in great detail, how this "other condition" is entered through ruptures in the
fabric of ordinary life. The ruptures that interested him particularly were the ecstasies
produced by music, sexuality and pure mathematics, all three of which he understood
as preliminaries to religious mysticism. ' Other mediations could easily be added, from
psychedelic drugs to a sense of humor (the latter is almost perfectly defined by the
aforementioned term Doppelboedigkeit). It is important, though, not to confuse the
mediation with the reality mediated; to do so would be falling into the generic fallacy
once again. Thus humor, for example, may be understood as a "signal of
transcendence," but it is not coterminous with transcendence: Laughing at a joke is not
in itself a religious experience. Mutatis mutandis, I believe, the same argument can be
made with regards to drugs.
The other reality posited by religious experience has, in all likelihood, been
perceived as a threat to ordinary social life from the begining. The universal ritual

'On the implications of Musil's novel, see Peter Berger (1970).

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
 132 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

response to it strongly argues for this: Ritual is universally a reaction to fear (very
probably a reaction with physiological roots). Be this as it may, society has always had
the problem of reintegrating and organizing these disruptions. Religious institutions
themselves are the most important mechanism for doing this. (Weber, by the way,
described this very clearly in his theory of the "routinization of charisma;" I think,
though, that the mechanism described occurs in a much broader area than the one
designated as "charismatic" by Weber). Religious institutions assign the potentially
disruptive manifestations of the other reality to carefully circumscribed times and
places in society. They "domesticate" the ecstasies, channel them into socially
acceptable and useful activity (such as moral conduct), and even manage to convert the
religious definitions of reality into legitimations of the sociopolitical order. The social
and psychological procedures of such "containment" are, I believe, an important topic
for the sociologies of knowledge and religion. The topic cannot be pursued here.
Suffice it to say that we have modern equivalents to the old exorcisms-psychiatry, for
one. (One additional observation here, to forestall predictable objections: What I have
just said about religious institutions in no way depreciates them, is not "anti-
institutional." I'm firmly convinced about the indispensability of institutions, in
religion as in every other area of human life. In other words: To describe the
"domestication" of religious experience is not necessarily to deplore it.)
While all this has been true for a very long time, and may even be rooted in the very
constitution of man, it takes on a special coloration in the context of secularization. It
is important for the thesis of this paper to look at this context.
Secularization can be defined as a shrinkage in the role of religion, both in social life
and in individual consciousness. Put in sociology-of-knowledge terms: Secularization
is a progressive loss of plausibility to religious views of reality. Now, there is a certain
ambiguity to this definition. It could imply two things: One, people are having fewer
religious experiences. Or, two, they still have these experiences, but, under social
pressure, they deny them. In the first case, modern consciousness would represent a
startling novum in human history. This has been maintained forcefully at least since
the Enlightenment, and it was stated most dramatically in Nietzsche's proclamation of
the "death of God." For better or for worse, the realm of transcendence would then
become closed to modern man. In the second case, we would have not so much the
disappearance of this type of experience, but rather its delegitimation. Religious
experience, so to speak, would be hidden in brown paper wrappers. Modern man
would then be not so much deprived of transcendence as dishonest in his reports about
it.
I suspect that both these things are true to some extent, but I am increasingly
inclined to think that the second is more important. I'm increasingly impressed by the
evidence that secularization is not quite tantamount to modernization-or,
alternatively, that modernization is not as complete as had been thought.
Secularization appears to be less far-reaching and less inexorable than many theories
of modern man had assumed. The Third World today is full of religious eruptions,
some of profound political significance. The evidence now emerging from the Soviet
Union (not only on the persistence of religion after a half-century of government
attempts to suppress it, but on the resurgence of religious impulses in the most unlikely
places) is downright astounding. But even in the western world there have been

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms
SECOND THOUGHTS ON DEFINING RELIGI6N 133

indications in recent years that (to paraphrase Mark Twain) the reports of God's
demise have been somewhat exaggerated. In this connection I have been much
impressed by the work in the sociology of religion of Andrew Greeley in this country
and of David Martin in England.
Perhaps the historians have been right all along, when they kept telling us social
scientists that the world hasn't changed all that much. Perhaps the peculiar
multidimensionality of religious experience is an anthropological constant, which will
reemerge over and over again, even if the "official" worldview of a society denies its
existence. Those of us who do, indeed, adhere to a religious interpretation of reality
will favor such thoughts. I do. But I am not saying that the scientific study of religion
should be based on this kind of philosophical affirmation. On the contrary, let me
reiterate that I firmly believe in the epistemologically neutral character of this
enterprise. It has been correctly said that the scientific study of religion must exhibit a
"methodological atheism." The adjective "methodological," though, should be
underlined. The scientific study of religion cannot base itself on any affirmation of the
ultimate truth claims of religion. But it must no more constitute itself on the basis of
atheism (that is, atheism tout court as against the aforementioned "methodological
atheism").
If my own sense of what is happening in the modern world is correct, then it is quite
possible that we may yet see more dramatic reversals of the process of secularization.
As we watch the stage of everyday life in the modern world, the action often seems to
take place on one level only. The "official" reality experts deny the rumblings that may
be heard from underneath-if necessary, they will sit on the trapdoor to make sure that
nothing can come up from the ominous cellar. My hunch is that their effort will fail:
The gods are very old and very powerful. But this is not what I have been arguing here.
My concern in this paper has been more modest: Whatever religious apparitions the
future may bring forth, it would be regrettable if the scientific study of religion were
systematically blinded to them by its own conceptual machinery.

REFERENCES

Bellah, Robert N. Geertz, Clifford

1964 "Religious evolution." American 1966 "Religion as a cultural system." Pp. 1-46
Sociological Review 29 (June): 258-374. in Michael Banton (ed.),
Berger, Peter L. Anthropological Approaches to the
1967 The Sacred Canopy. Garden City, N.Y.: Study of Religion. New York: Praeger.
Doubleday. Luckmann, Thomas
1970 "The problem of multiple 1967 The Invisible Religion. New York:
realities-Alfred Schutz and Robert Macmillan.
Musil." Pp. 213ff in Maurice Natanson Musil, Robert
(ed.), Phenomenology and Social 1965 The Man Without Qualities. New York:
Reality. Essays in Memory of Alfred Putnam.
Schutz. The Hague: Nijhoff.

This content downloaded from 130.113.111.210 on Thu, 19 May 2016 10:42:42 UTC
All use subject to http://about.jstor.org/terms