THE PROBLEM OF DEFINING RELIGION

J. Kuruvachira

Introduction
Defining religion is seen as a major problem in contemporary religious studies. In fact, today, a
commonly accepted definition of religion hardly exits among scholars. Any attempt to define religion
would imply retaining some aspects and omitting others. The difficulty in defining has only increased
in the wake of our knowledge that there is almost an infinite variety of religions, each proposing
worldviews that considerably differ from one another, and at times even conflict with each other.
Religious scholars have put forward a wide variety of definitions. It might be stated that religion is such
a complex and diverse cultural phenomenon that reducing it to any single definition would either fail to
capture what it really is, or merely misrepresent it. Some do not define religion, arguing that the
definitions one begins with often influences the course of research in decisive ways. Others put forward
definitions but fall short of universal acceptance. Yet religion is a major factor in human life and it is
likely to remain so for the foreseeable future. Hence, one needs to know what religion is and what its
essential elements are. This means that attention to the problem of defining religion is a fundamental
prerequisite to the study of religion. In this article I will briefly examine the concept of ‘religion’,
problem of defining religion, major types of definitions with a critique of the functional definition of
religion and definition of religion by essential elements, and conclude by arguing that a complementary
approach to the definition of religion might be more suited to a complex subject like ‘religion’ than a
compartmentalised approach.

1. The concept ‘religion’

Many people falsely assume that ‘religion’ is a simple term whose ‘essence’ can be easily
apprehended. The concept ‘religion’ is modern, and its use dates back to the 16th century, and is a
creation of Western scholars. It is sometimes used interchangeably with ‘faith’ to indicate a belief
system that has a communitarian aspect. In dictionaries, ‘religion’ is often defined as a specific system
of belief, worship or conduct that prescribes certain responses to the existence and nature of God or
some supernatural power or transcendent reality.
The origin of the term ‘religion’ is uncertain. Some say that it is derived from the Latin religio
(‘attach’ or ‘conserve’ or ‘moral bond’). According to Cicero (106 B.C.E- 43 C.E) it comes from
relegere (‘to re-read’ to mean ‘carefully consider things related to the cult of the gods’). Lactancius, a
Christian author of the fourth century, sees its origin in the religare (‘connect’ or ‘join’) to refer to the
bond of piety which binds one to God.1
The meaning of the word ‘religion’ has changed over the last five centuries. In the 16th century it
indicated the institutional life of the Christian Church. The faith practices of the non-Christians were
considered either idolatry or ‘fashions’.2 Joseph de Acosta (1540-1600), a Spanish Jesuit, used the word
religion’ to mean ‘a belief system that results in a ceremonial behaviour’. In the 18th century Samuel
Johnson in his Dictionary of the English Language (1755) used ‘religion’ as ‘virtue’. The first edition
of Encyclopaedia Britannica (1771) employed the term ‘religion’ as a substitute for ‘theology’. Ulrich
Zwingli (1484-1531) and John Calvin (1509-1564) used ‘religion’ to mean ‘piety’.3 In the Romantic
era (18th and 19th centuries) ‘religion’ came to refer to personal attitudes, and became synonymous with
‘faith’. As knowledgement of other religions increased from the late 18th century, the word ‘religion’

1 Cfr. U. Berner, “Religione”, Nuovo Dizionario delle Religioni, Hans Waldenfels (ed.), Milan, Edizione Pauline, 1993,
756-757; Paolo Grassi, “Religione”, in Enciclopedia Filosofica, Vol.10, Milano, Bompiani, Fondazione Centro Studi
filosofici di Gallarate, 2006, 9578-9579.
2 Cfr. Charles B. Jones, Introduction to the Study of Religion, Parts I & II. Virginia, The Teaching Company, 2007, 5.
3 Cfr. Jonathan Z. Smith, “Religion, Religions, Religious”, in Mark C. Taylor (ed.), Critical Terms for Religious Studies,
Chicago, The University of Chicago Press, 1998, 270-271.
1
came to indicate a category of which other religions were equal members.4 Today many scholars are
inclined to use the term ‘tradition’ for what we commonly understand by ‘religion’. Thus they speak of
Christian tradition, Buddhist tradition, Islamic tradition, Judaic tradition, etc.
The word ‘religion’ can have different shades of meaning across diverse cultural contexts. Hence the
use of cross-cultural concepts can become advantageous because of their remarkable power to make the
nuances clearer, richer and more accurate in languages other than European. For example, some
alternative concepts for religion in non-European cultures are: dharma (teaching, law), darsana
(worldview), marga (way) in Sanskrit, and chiao (teaching) and Tao (way) in Chinese.5
Some scholars consider religion as a species of the genus ‘worldviews’ which includes also secular
ideologies (like Atheism, Communism, Existentialism, Humanism, Marxism, Maoism, etc). According
to them, those systems which are conventionally and traditionally called religions are only one species
of the worldviews, and the rest could be called non-religious worldviews.6
It has been argued by some that there is no such thing as ‘religion’ per se, only culture, and the
various cultural manifestations, and that some aspects of culture have been arbitrarily singled out and
grouped together, and given the label ‘religion’. According to them ‘religion’ exists merely in the
minds of scholars studying culture, and has no independent existence of its own.7
There are scholars who argue that the term ‘religion’ should be used in the plural because all
religions are particular and there exists no religion in the abstract. Ninian Smart says: “Perhaps there is
no such thing as religion in general – all that we meet are particular religions.”8 However, there are
others who admit a plurality of religions but also see certain common elements among them which
make it possible to use the term ‘religion’ in the singular. For instance, the Jesuit scholar Giovanni
Magnani holds that there is some common and universal phenomenon which unites the different
religions, and therefore the term can be used in the singular.9 Friedrich Schleiermacher adds that
religion is infinite in nature and in the different manifestations of religion one should discover the
universal element, namely, ‘religion’. He says: “in the religions, you are to discover religion”.10 Then
there are others who maintains that ‘religion’ is a univocal term11 and can be predicated to any
experience of the transcendent or something analogous to it in any culture.
These conflicting views about ‘religion’ only reveals that it is a very complex concept and there are
no easy answers as to what does and does not qualify as ‘religion’. There are people who think that
religion is all very simple and obvious, but they merely betray a superficial and simplistic familiarity
with the concept and fail to capture its multifaceted aspects and the difficulty in articulating what it
really is. This has even led some scholars to drop the word ‘religion’ altogether. Wilfred Cantwell

4 Cfr. Charles B. Jones, Introduction to the Study of Religion, 5.

5 Cfr. Ninian Smart, “Religion”, in Alan Richardson and John Bowden (eds.), A New Dictionary of Christian Theology,
London, SCM Press, 1984, 498.
6 A major exponent of this a view is Ninian Smart. Cfr. Ninian Smart, Buddhism and Christianity: Rivals and Allies,
Honolulu, Hawaii University Press, 1993, 6.
7 This idea has been put forward by Jonathan Z. Smith in his Imagining Religion (1998)
http://atheism.about.com/od/weeklyquotes/a/smith01.htm. Accessed on 7 March 2011; Henry Goldschmidt, “Religion,
Reductionism, and the Godly Soul: Lubavitch Hasidic Jewishness and the Limits of Classification of Thought”, Journal of
the American Academy of Religion, Vol. 77 (2009), No.3, 549.
8 Ninian Smart, “What is religion?”, in Ninian Smart and Donald Horder (eds.), New Movements in Religious
Education, London, Temple Smith, 1975, 13.
9 Cfr. Giovanni Magnani, Metodologia Fenomenologica Interdisciplinare delle Science della Religione, Roma,
Università Gregoriana, 1996, 51.
10 Friedrich Schleiermacher, On Religion. Speeches to Its Cultures Despisers, Richard Crouter (trs.), Cambridge,
Cambridge University Press 1988, 190.
11 Cfr. Adriano Alessi, Sui Sentieri del Sacro. Introduzione alla filosofia della religione, Roma, LAS, seconda edizione,
2009, 314.
2
Smith, for instance, uses the term ‘faith’ instead of ‘religion. Thus he speaks of ‘the Christian faith’,
‘the Jewish faith’, ‘the Buddhist faith’, ‘the Hindu faith’, etc.12

2. The problem of defining religion

On the question of defining religion, scholars can be broadly divided into two categories, namely,
those who deny the possibility of defining religion, and those who advocate its possibility. The former
group claims, for instance: 1) the definitions of religion put forward in the past have been repeatedly
ignored by succeeding generations as inadequate. So it is better to learn from experience and accept
that religion is not definable; 2) from a linguistic point of view, the quest for a conceptual definition of
religion is to be dismissed, because religion makes sense only in the context of our ordinary language;
3) all that we know are particular religions and this means that there is no single essence to religion and
hence no unified definition of religion is possible.13
Ninian Smart argues that: 1) every definition of religion involves a particular theory about a religion,
which will consequently rule out some phenomena as not really religious. This implies an implicit
theory of what really religion is, and it means begging the question; 2) there can be as many definitions
of religion as there are disciplines, and the term ‘religion’ can have a variety of connotations in the
minds of those who employ the concept; 3) an insiders’ view of religion can differ from an outsiders’;
4) the study of religion is without clear boundaries and so it is not possible to provide a clear-cut
definition of religion that would distinguish it from and secular ideologies which may not be normally
placed on a par with the traditional religions.14
Zygmut Bauman claims that defining religion amounts to replacing one ineffable with another, to the
substitution of the incomprehensible for the unknown.15 Heinrich von Stietencron maintains that all
religions are constantly in evolution and none of them is static. They change, branch off, and transform,
sometimes almost beyond recognition.16 Consequently this constantly changing character of religion
adds to the problems of defining religion. Hence the attempt to formulate a universally accepted
definition of religion may not be easy, if not impossible.
But those who advocate the possibility of defining religion admit the corrigibility of definitions of
religion and its plurality. Definitions are meant to be tools for bringing order and discipline and for
providing scope and precision to our conceptual world of experiences. Hence religion can be defined
by taking into consideration certain common characteristics or essential elements. In addition, although
defining religion is difficult, the methodology of the study of religion presupposes a formulation of at
least a tentative definition precisely because one should know, at least in some degree, the nature of the
subject under investigation or what counts as religion. Therefore, we need to define ‘religion’ for the
purpose of clarity and to guide those who study religion. Besides, the fact that scholars have attempted
definitions of religion in the past only shows that it is possible to define it, though with different
degrees of acceptance. For instance, as early as 1912 James H. Leuba provided 48 definitions of
religion with his critical comments on each to indicate that the attempt to define religion is an
enterprise that is practically useless.17 But some scholars have argued that Leuba has shown that we can
actually define religion at least in 48 ways with a greater or lesser success.18 Today any scholar of
religion knows that there is almost an infinite variety of definition of religion available in books of

12 Cfr. Wilfred Cantwell Smith, Patterns of Faith Around the World, Oxford, Oneworld, 1998, 8-12.
13 Cfr. Frederic Ferré., “The Definition of Religion”, Journal of the American Academy of Religion 38 (1970), 3-6.
14 Cfr. Ninian Smart, “What is religion?”, 13.
15 Cfr. Zygmut Bauman, “Postmodern Religion?”, in Paul Heelas (ed.), Religion, Modernity an Postmodernity, Oxford,
Blackwell, 1998, 55.
16 Cfr. Heinrich von Stietencron, “Der Begriff der Religion in der Religionswissenschaft”, in Walter Kerber (ed.), Der
Begriff der Religion, München, Kindt 1993, 117.
17 Cfr. James H. Leuba, A Psychological Study of Religion, New York, Macmillan, 1912, 339-361.
18. Cfr. Jonathan Z. Smith, “Religion, Religions, Religious”, 281.
3
philosophy, theology, history of religions, sociology, anthropology, psychology, jurisprudence, etc.
Hence defining religion is possible.

2.1 Types of definitions

There are different approaches to definition of religion, and are generally called by such names as,
essential-intuitive definition, lexical definition, functional definition and real definition.

2.1.1 Essential-intuitive definitions

Those who advocate the essential-intuitive definitions of religion are actually refusing to put a
definition forward. They follow the policy of ‘I know religion when I see it’ approach.19 This method is
widely used without raising basic questions about definition and proceeds as though the word ‘religion’
is simple and unambiguous, that everyone knows what is meant by the word, and that the reality or the
essence of which the term refers is intuitively identifiable. If the question of definition arises at all
(often it does not arise), it arises at the end of the study or at the end of one’s career. But critics argue
that ‘religion’ is a far too complex an entity to be left to the ‘intuitive’ method. Besides, the study of
religion is a scholarly pursuit where one needs to clarify a basic term like ‘religion.20
2.1.2 Lexical definitions
Lexical definitions describe how words have been used historically. This is what dictionaries
provide.21 Lexical definitions provide us evidence of how a term has been used at a given time by a
person or group. But they can be ambiguous since the past usage of a word can differ in a present
context.22 This is especially evident from the use of the word ‘religion’ in the course of history.
2.1.3 Functional definitions
Functional definitions, which are semi-arbitrary definitions, specify how a given term is used in a
finite context.23 It is called functional in terms of its place in the social or psychological or any other
particular system. It deals with what religion does for a person or society or both. In it the ultimate truth
claims that are central to religion are left out. The functional definitions are nominal in character, and
since nominal it also lacks truth-value. Since it is a decision to use a word in certain way one cannot
ask if the definition is true or false, nor is it appropriate to test its truth by examining data. But
functional definitions lend precision to a study by removing ambiguity in words.24
2.1.4 Real definitions
Real definitions are those that try to capture the essence of reality.25 They are also called definition
by ‘content’ or ‘essence’ or ‘substance’ (hence substantive definition). The search for the essence is an
attempt to find an identity in the numerous applications of a word. Religion is such a word, and the
essence of religion is what religions have in common. A real definition is not only true or false but
presumably come at the end of a study.26 However, the problem with real definitions is centred around
what the definer takes to be the ‘essential nature’ of some entity.27

19 Cfr. Charles B. Jones, Introduction to the Study of Religion, 5.

20 Cfr. Robert D. Baird, Category Formation and the History of Religions, The Hague, Mouton, 1971, 2-5.
21 Cfr. Charles B. Jones, Introduction to the Study of Religion, 6.
22 Cfr. Robert D. Baird, Category Formation and the History of Religions, 10-11.
23 Cfr. Charles B. Jones, Introduction to the Study of Religion, 6.
24 Cfr. Robert D. Baird, Category Formation and the History of Religions, 6-8.
25 Cfr. Charles B. Jones, Introduction to the Study of Religion, 6.
26 Cfr. Robert D. Baird, Category Formation and the History of Religions, 11-14.
27 Cfr. Melford E. Spiro, “Religion: Problems of Definition and Explanation”, in Michael Banton (ed.), Anthropological
Approaches to the Study of Religion, London, Tavistock Publications, 1996, 86.
4
 We shall analyse below the merits and demerits of the functional and real definitions as they are the
kinds of definition we frequently encounter.

2.2 Some conditions for defining religion

Over the centuries, scholars have offered numerous definitions of religion, but in practice none of
them is satisfactory. In some cases the definitions are too narrow and describe religion in terms of the
religious beliefs of the author, or those of his or her culture, and tend to exclude the religious beliefs of
other cultures. Many definitions focus too narrowly and highlight only some selected aspects of
religion, and consequently they tend to exclude those religions that do not fit well in their scheme. In
other cases the definitions are too broad and embraces a penumbra of worldviews which are clearly
secular and at times even opposed to religion.
In order to arrive at a good definition of religion we need to keep in mind certain conditions. First,
the definition should be free from criteria which are culturally determined. In other words, we may not
look at two or three cultures and base our definition solely on their religious structures. For instance,
defining religion in terms of reference to supernatural being (beings) who is responsible for the creation
of the universe and for its continuing operation. If we define religion in this way Buddhism, Jainism,
Confucianism and so on, with good claims to be ‘religions’, can never be regarded as religions as they
do not speak of a God per se who is the creator and maintainer of the universe.
Some definitions of religions have the danger of applying prior ideological constraints and cutting
off other avenues of inquiry. An example of this is Emile Durkheim’s definition of religion as ‘an
eminently social thing’.28 Some other case are, Ludwig Feurbach’s view that religious realities are
‘projections of human qualities’, Frederic Nietzsche’s assertion that religion is a manipulative device
that merely serves human power-interests29, Karl Marx’s definition of religion as ‘opium of the
people’, Sigmund Freud description of religion as a ‘universal obsession neurosis of humanity’ 30 and
Richard Dawkins’s view that ‘religions are mind viruses’31, to mention a few.
Since religion is a complex and diverse cultural phenomenon, a definition of it must be sufficiently
broad to apply it equally to any culture. At the same time it must be narrow enough not to be placed in
par with such secular ideologies as Marxism, Communism, Nazism, Existentialism, Atheism,
Humanism, etc., which are often in conflict with traditional religions.
Religion being an existential reality, a definition of religion has to take into consideration the
parameters of context and plurality. Context here implies knowledge about religions in their concrete,
historical manifestations. Plurality signifies considering religions cross-culturally.
Therefore, the conditions for a good definition of religion are that they should be narrow enough to
afford a clear distinction between what is ‘religion’ and ‘non-religion’ and broad enough to be
applicable to any culture so that no religion, recognised as such, is left out. A definition should be made
in a state of freedom from prior constraints, be it ideological or personal. A good definition should
indicate how religion affects the life of individual persons and society and gives them meaning. Finally,
a definition should deal with those fundamentals of religion that remain stable in the midst of the
changes that religions undergo. In other words, a good definition should contain the universal or
essential elements found in every religion.

28 Cfr.Emile Durkheim, The Elementary Forms of the Religious Life, tr. Joseph Ward Swain, New York, The Macmillan
Company, 1915, 10.
29 Cfr. Anthony C. Thiselton, A Concise Encyclopaedia o the Philosophy of Religion, Oxford, Oneworld, 2006, 262.
30 Sigmund Freud, The Future of an Illusion, James Strachey (trs. and ed.), New York, W.W. Norton and Company,
1961, 55.
31 http://en.wikipedia.org/wiki/The_God_Delusion. Accessed on 11 March 2011. Richard Dawkins (b.1941-) is a British
ethnologist and evolutionary biologist who openly states that he loathes religion. His recent book The God Delusion (2006)
is not only a defense of atheism, but also goes on the offensive against religion. He sees religion as subverting science,
fostering fanaticism, encouraging bigotry against homosexuals, and influencing society in other negative ways. Ibid.
5
2.3 The classical dilemma in defining religion
If we maintain that we need to define religion we are faced with a classical dilemma: should the
definition precede the analysis of the data about religion, or should it follow the analysis? Some
scholars assert that unless we start out with a definition of religion and specify what we are looking for,
there is no way to separate religion from non-religion. Therefore, we must have some preliminary
concept of ‘religion’ at the start, or we might end up studying something other than religion, like
culture, or theology, or psychology or sociology or anthropology or something else. Others retort that if
one defines the subject beforehand, he or she might distort the data owing to preconceptions, or at very
least he or she might miss some data which are really relevant, but which have been mistakenly
excluded from the outset. Hence they maintain that after having studied the data about religion one can
define what is religion. Max Weber was one who argued that if we attempt to define religion at all, it
must be done at the conclusion of the study.32 The Italian philosopher of religion Adriano Alessi holds
a similar view.33
But we propose a way out of this dilemma. We can divide the definitions of religion into two
categories, each serving a special function and only a function. We could call them ‘preliminary
definitions’ and ‘final definitions’.34 We could start with a number of elementary preliminary
definitions for determining the subject matter of the study, and conclude with some ‘final definitions’,
either by refining some of the existing definitions or by creating new ones altogether. The preliminary
definitions should be several. It is best done by referring to the definitions of some of the well known
scholars of religion over a wide timeframe, from different disciplines, and from cross cultural contexts,
and then briefly criticism them from one’s own experience and knowledge of religions, in order to
discover what are the elements that are positive and negative in them. At this stage it is also important
that different categories of definitions are provided, such as, essential-intuitive, lexical, functional,
substantive, etc. At the final stage too it is good to furnish more than one ‘final definition’ of religion so
that one is made aware that no definition is absolute and that each definition is open to further
refinement. The plurality of definitions at the final stage will also allow one to choose from several
options according to one’s preference or discard them altogether for a new definition of one’s own
formulation, because as Peter Berger says, “definitions are matters of taste and thus fall under the
maxim de gustibus”.35

2.4 An assessment of the functional definition of religion

A functional definition of religion, as we have seen, is one that is based on the supposed impact or
functions of religion in the life of individuals or on society or both. Those who advocate the functional
definition argue that it can be applied equally well to theistic and non-theistic belief-systems, and to
societies which recognize a supernatural order, and to those which make no such recognition. They
further claim that it can account for historical change and discontinuity in religions better than any
definition by essential elements. They also insist that it can be empirically verified, so that it can be
proven true or false.36
A classical example of a functionalistic explanation of religion is Emile Durkheim's description of
religion as ‘an eminently social thing’ because of its integrative or socializing power in society.
Another instance is Max Weber’s interpretation of religion as ‘source of social action’. But these
definitions are reductionist because they do not accept that religion is a valid reality in its own right,

32 Cfr. Max Weber, Sociology of Religion, Boston, Beacon Press, 1963, 1.

33 Cfr. Adriano Alessi, Sui Sentieri del Sacro. Introduzione alla filosofia della religione, 306-310.
34 In making this distinction, I have been influenced to some extent by Russell Kirkland’s distinction between ‘field-
delimitation’ or delimitative definition, and ‘inductive definition’ or determinative definition. But I have tried to follow a
simpler approach than his. Cfr. http://kirkland.myweb.uga.edu/rk/pdf/guides/RELDEFINE.pdf. Accessed on 13 March 2011
35 Peter L. Berger, The Sacred Canopy. Elements of Sociological Theory of Religion, New York, Doubleday, 1967, 177.
36 Cfr. http://kirkland.myweb.uga.edu/rk/pdf/guides/RELDEFINE.pdf. Accessed on 13 March 2011.
6
instead see it as a function of social process that should be reduced to sociological explanations.
Besides, both scholars reject any reference to a supernatural order. Other sociologists of religion have
correctly pointed out that even if religion does serve an integrative function and source of social action
in some societies, it is wrong to assume that such a function accounts for the existence of religion in
general.
Another example of a functional definition is Clifford Geertz’s description of religion as a ‘system
of symbols’.37 Here, by symbols he means that which synthesise the character and quality of life, its
moral and aesthetic style and mood. But in essence this description emphasizes only the ethical,
cultural, pragmatic and empirical aspect of religion and omits any reference to the supernatural. He
forgets the fact that some symbolic expressions can also be religious acts that point to a transcendental
reality or supernatural order by virtue of the intentionality of the subject. In fact, the primary purpose
of the symbols used in religion is to point to transcendental realities as there is no possibility of direct
access to them.
A psychologist who approached religion from the functional point of view is Carl Gustav Jung. He
paid attention to the effects of religion on human beings rather than the reality of its referents. In fact,
unlike Sigmund Freud, he considered religion as a necessary component of mental health. But he did
not care to ask whether the referents of religion really existed, rather merely noted the effect of the
experience the objects of religion on people.
There are some definitions of religion which emphasise the power of religion to answer certain basic
questions of life and resolve its problems, whether apparent or real. Such definitions usually consider
religion in psychological, or existential terms, or both. The basic premise of this view of religion is that
human beings face certain ‘basic problems’ in life (like sickness, suffering, death, injustice, natural
calamities, etc), which cannot be escaped. They produce frustration, dissatisfaction, and anxiety, for
which people need some solution, if life is to be bearable. Many see religion as a means of coping with
the basic problems of human existence.38 This again can be considered as a functional way of
conceiving religion. But this ‘meaning providing aspect’ does not exhaust the nature and purpose of
religion, though it can be considered as one of the elements of it.
The functional definition of religion has also other disadvantages. A functionalistic definition of
religion is an explanation that there is nothing to religion (or a specific religion) beyond the social or
psychological or anthropological or cultural functions which it performs. A definition of religion which
takes into account solely its functions inherently tends toward reductionism. A functional definition
explains what religion does (or can do), for society, for individuals or for both, and claims to have
explained what religion is all about. But religion, which is a multifaceted reality, cannot be reduced to
some functions alone, and a functional definition cannot pretend to be the sole explanation of religion.
Another problem with the functional definitions is that the essential transcendence of the object
intended by religious acts (transcendent intentionality) is totally ignored by functional definitions, and
consequently religion is reduced to culture or sociology or anthropology or psychology or something
similar. The functional definitions being empirical in nature, totally sidesteps the ontological question
of reference to transcendental realities, which are so fundamental in religion. Unless the functional
definitions take into account the irreducible element of transcendence in religion, the reductionist
tendency cannot be overcome. Peter Berger has observed that in the 19th century it was common to

37 He defines 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." Clifford Geertz, “Religion as a Cultural system”,
in Michael Banton (ed.), Anthropological Approaches to the Study of Religion, London, Tavistock Publications, 1996, 4.
38 Cfr. http://kirkland.myweb.uga.edu/rk/pdf/guides/RELDEFINE.pdf. Accessed on 13 March 2011.
7
have definitions of religion which implied a reductionist philosophy of religion, and he called this
method ‘assassination through definition’.39
Another criticism against the functional definition is that anything which performs the prescribed
function thereby becomes ‘religion’. Such a definition does not permit any clear boundary separating
religion from non-religion. A typical case of this is Paul Tillich’s definition of religion as ‘Ultimate
concern’. Anything can be an ultimate concern for a person or society. It does not mean that every
‘ultimate concern’ is religion. Therefore, no purely functional definition of religion can be considered
as a satisfactory definition of religion.

2.5 An assessment of definition of religion by essential elements

Another definition of religion which we frequently encounter, is definition by the supposed
‘essential elements’ or ‘common content’ or ‘substance’. Some theorists strongly prefer this way of
defining religion and consider it as closely resembling the commonsense approach. 40 A definition by
essential elements is concerned with what religion is. In order to discover the common content or
essential elements of religion, one has to employ the method of cross-cultural comparison, analysis,
analogy, induction, deduction and intuition. The purpose of it is to identify and separate the ‘essence’
of religion from the ‘accidents’.
According to Peter Berger, a definition by essential elements (which he calls ‘substantive
definition’), generally includes only such meanings and meaning-complexes as refer to transcendent
entities in the conventional sense, such as, God, gods, supernatural beings and worlds.41 Some
examples of it are, E.B. Tylor’s definition of religion in terms of supernatural or superhuman beings,
Rudolf Otto’s and Joaquim Wach’s definition by reference to ‘holy’, Mircea Eliade’s definition by
reference to ‘Sacred’, Gerardus van der Leeuw’s definition by reference to ‘Power’, etc. But in recent
years several scholars have objected to the use of the above terms, since they consider them vague, and
cannot be empirically verified.
The distinction between ‘essential’ and the ‘accident’ is central to the definition by content or
substance, and precisely around this issue conflicts have taken place.42. Some say that determining what
is ‘essence’ and what is ‘accident’ already implies an evaluation. Then there are others who argue that
there is no such thing as common content in religions. For example, according to Ninian Smart there is
no common core or essence in religions. He says: “let us look first at Christianity and then at
Buddhism, a good contrast of styles and content. One cannot expect a single core of content in the
religions”43. Mark Owen Webb holds a similar view and argues that there is no essence to religion and
what we call religion is a heterogeneous collection of practices which are so different from one
another.44 Hence the search for a common essence will not be useful, and consequently no single
definition of religion in terms of essence will be possible.
However, it is to be remembered that the use of analogy is important in identifying the essential
elements of religions. If we take certain key concepts of religion as univocal, certainly we may not find
anything common among them. Perhaps the scholars who refuse to see any common element in
religions are studying the religious data using a methodology where there is no place for analogy. And

39 Cfr. Peter Berger L., "Some Second Thoughts on Substantive versus Functional Definitions of Religion," Journal for
the Scientific Study of Religion, Vol. 13 (1974), 126.
40 Cfr. Daniel L. Pals, Eight Theories of Religion, Oxford, Oxford University Press, 2 nd edition, 2006, 13.
41 Cfr. Peter L. Berger, "Some Second Thoughts on Substantive versus Functional Definitions of
Religion", 127-128.
42 Cfr. Karel Dobbelaere, Jan Lauwers, “Defining Religion – A Sociological Critique”, Social Compass, Vol.20 (1973-
1974), 549.
43 Ninian Smart “What is Religion?”, 13-14.
44 Cfr.Mark Owen Webb, “An Eliminativist Theory of Religion”, Sophia, Vol. 48 (2009), 36.
8
yet if we apply analogy as part of the methodology of the study of religion, it may not be difficult to
discover certain common elements in them.
Another problem with the scholars who object to definition by essential elements is that their study
of religion remains at the level of the phenomena, and they fail to make a philosophical reflection on
the phenomenological data. This is once again a methodological deficiency. Unless the
phenomenological data is subject to philosophical reflection through rigorous use of comparison,
analysis, deduction, induction and intellectual intuition, one cannot discover any common element in
religions. For instance, religions may refer to a transcendent reality using different names, such as
Brahman, Trinity, Nirvana, Emptiness, state of isolation, etc. When we analyse these concepts from a
philosophical perspective it is not difficult to conclude that they all refer to a transcendent reality,
understood analogically.
Those who study religion must admit that not all the essential elements of religions are empirically
verifiable, though some of them can be, such as rituals, myths, narratives, institutions, religious
authority, religious community, ethical code, ability of religion to help people cope with life, etc. But at
the same time religious symbols and ritual acts have an intentionality that point to certain transcendent
realties, and which are beyond the grasp of those who employ the empirical method.
Certain essential elements of religion can be reached only through a philosophical reflection on the
religious data. Such an approach is justified because the study of religion belong to humanities and not
to science proper where the method employed is empirical. Strictly speaking the study of religion does
not fall within the field of science. Science has nothing to say about assertions that are beyond the
empirical. No one can prove scientifically that the supernatural beings or transcendental states about
which religious people speak do not in fact exist. As non-empirical entities, their existence can be
neither proven nor disproven by the method of science. Since religion, by any reasonable definition,
contains non-empirical elements, it is clearly an unfit object for application of the scientific method.45
Hence there is no a priori reason to deny the validity of definition of religion by essential elements.

3. Need for a complementary approach to the definition of religion

We have seen that neither a functional definition of religion nor a definition by essential elements is
free from limitations. This means that neither approach can be considered as absolutely perfect or fully
satisfactory. Both can have ideological intent, and consequently both can be limited. In order to have a
more balanced approach to the study of religion we might need definitions that take into consideration
both the functional and the essential aspects of religion. Hence we propose a complementary approach
to the definition of religion, i.e., a definition which will be a synthesis of both kinds of definition. A
complementary approach can also deal better with the complexity of the concept of religion than a
single kind of definition, be it a functional definition or a definition by essential elements.
A complementary approach to the definition of religion has the necessary ingredients to be a
balanced one because it is related to the very nature of religion itself, which is complex and multi-
faceted. A religion has immanent and transcendent, natural and supernatural, observable and non-
observable, phenomenological and metaphysical, empirical and speculative, practical and intentional
aspects. Unless these aspects are taken into consideration in defining religion, our understanding of
religion can be very limited and partial.
Again, since religions have humanist, psychological, social, cultural, economic, political,
anthropological and transcendental orientations, a good definition of religion should deal also with all
these aspects as they can complement each other and help one understand what religion is. Such an
approach will also indicate that the study of religion is much more than a purely speculative or
pragmatic enterprise.

45 Cfr. http://kirkland.myweb.uga.edu/rk/pdf/guides/RELDEFINE.pdf. Accessed on 13 March 2011.

9
 By advocating a complementary approach to the definition of religion by integrating the functional
definition and definition by common essence, we are not admitting that the two types of definitions
have the same value. Their value will be determined by the particular disciplines that use the
definitions. For the discipline of philosophy, a definition especially by essential elements or common
content might be more valuable and immediately relevant than a functional definition; for a
psychologist a definition of religion that highlight the ‘meaning-giving’ aspect of religion might be
more valuable than a definition by reference to a sociological function of religion; for a sociologist a
definition of religion by referring to a sociological aspect of religion might be of greater relevance than
a definition by essential elements, and so on. But at the same time, if these disciplines remain closed in
on themselves, they may not be able to have holistic view of religion.
A complementary approach to the definition of religion necessarily calls for an inter-disciplinary
approach to the study of religion, where the findings of such disciplines as, philosophy, theology,
sociology, anthropology, psychology, politics, economy, phenomenology, etc, are utilised and
harmoniously integrated into the understanding of what religion is and what it does for individuals and
societies. It will also demand the avoidance of an exclusivist approach in defining religion.
If we follow a complementary approach, there is no need to contrast the functional definition of
religion with those by reference to the essential elements and vice versa. Each type of definition can
shed some light in understating (Vestehen) religion. This means that the ‘preliminary definitions’, about
which we spoke above, should consist of both the kinds of definitions, and the author of the ‘final
definitions’ should manifest his or her capacity to formulate definitions of religion by reference to
different disciplines.

Conclusion
We conclude by affirming that defining religion is possible and necessary for the study of the
religious data. Definitions of religion can serve to stimulate the understanding of the nature of religion.
Religion is a complex phenomenon, and as such no single definition can capture all its aspects. Hence
there is no point in advancing a single definition of religion hoping that it would meet with universal
acceptance or be rated as the best by scholars. But this need not deter us from attempting definitions of
religion. But in defining religion one should keep aloof from ideological standpoints, and personal like
and dislikes. One should also make a clear distinction between a definition of religion and an
explanation of religion. Since religions are dynamic in nature and constantly evolving, the definitions
of religion must also become an on-going process. One should constantly re-think the fundamental
elements of one’s definition and refine them in accordance with the changed views in different
disciplines. As Peter Berger says, definitions are always ad hoc constructions and have a definitive
cognitive purpose.46 Finally we maintain that a complementary approach to defining religion is better
than a functional definition or definition by essential elements taken independently. Such an approach
would imply that the study of religion should be interdisciplinary.

46Cfr. Peter Berger L. "Some Second Thoughts on Substantive versus Functional Definitions of
Religion", 127.
10