We need working definitions as part of our methodology. Definitions are not final statements of absolute truth, but part of a working methodology which make explicit what is and is not included, what is the focus of our field, what distinguishes it from neighbouring fields, and so on. Inadequacies can be argued about and corrected. Thus my definition of cultural studies leaves open for discussion whether or not it is the best formulation, but I would say now that I believe social and cultural anthropology provides us with an important part of any new framework, as do literary criticism and history (2).

But in my view we have to develop a new paradigm for religious studies which more accurately represents what the majority of scholars working in religion departments are actually studying. This is because the actual usage of the word Religion, religions and world religions by scholars in their

FALSAFEH   Vol. 37, No. 3,  Autumn 2009, pp. 21-36

publications reveals fuzziness, ambiguity and contradictoriness.
There is an understandable instinct for western scholars who come from a monotheistic tradition to equate religion with belief in the transcendent or the supernatural, and soteriological doctrines concerning the salvation of the individual. Analysis shows that religion is often implicitly or explicitly defined in this way. This is particularly the case when borderline cases occur, such as magic, witchcraft, 'non-religious' ideologies such as Confucianism, Maoism or Marxism, politics, economics and other institutions. Borderline cases often place an obligation on an author to explain how s/he is and is not using key terms.

Yet at the same time the word religion is actually used in so many different contexts that it is too indefinite to have any analytical value. It is used to talk about salvation doctrines, belief in the 'supernatural', belief in ghosts, ancestor worship, ultimate values such as the family or egalitarianism or deference or hierarchy; or ideology in general; buying a Christmas cake or an amulet; supernatural technology such as possession and exorcism; Emperor worship, caste ritual, gift exchange, and so on.

Sometimes authors write about the religion of a society which they have studied and simultaneously admit that the word has no referent. For example, Cooper (1988) (3) in an interesting article titled 'North American Traditional Religion', feels obliged to sprinkle his article with the 'R' word even though he has said at the beginning:

"No tribe has a word for 'religion' as a separate sphere of existence. Religion permeates the whole of life, including economic activities, arts, crafts and ways of living. This is particularly true of nature, with which native Americans have traditionally a close and sacred relationship. Animals, birds, natural phenomena, even the land itself, have religious significance to native Americans: all are involved in a web of reciprocal relationships, which are sustained through behaviour and ritual in a state of harmony. Distinctions between natural and supernatural are often difficult to make when assessing native American concepts." (873-4)

When the author points out that 'religion' permeates the whole of life, the reader can wonder what is the difference between saying that and saying that the concept has no distinct meaning,
because nothing is picked out by it. One thinks of Anthony Flew's parable of the invisible gardener, but here the term which makes no difference is 'religion' not 'God'.

Somebody might reasonably claim against me that, where there is no 'religion' existing as a distinct institution, it might still be legitimate to say that there is a 'religious aspect to existence'. But what does this consist of in the American Indian case, as presented by Cooper? It consists of sacred relationships, the sacred being what is most deeply valued by the social group. In order to draw close to these values and their institutionalisation we have to look at the whole 'web of reciprocal relationships' as Cooper has put it. So the meaning of 'religion' here seems to be what this group of people value most deeply, and the institutions and relationships which reveal those values. How these values are related to the legitimation of power in American Indian societies is not addressed by Cooper.

Let us turn to Hinduism, which in many books over the last twenty or so years has been described as a Religion, or as a World Religion, or less often as several religions. Hinnells and Sharpe, in an early (1972) but influential book Hinduism, acknowledge the problem of caste in this way:

"A Hindu is a Hindu not because he accepts certain doctrines or philosophies, but because he is a member of a caste." (1972:6)

Given the actual contents of the book, this is a surprising admission.

There are less than three pages on caste. The section on caste is no longer than the average length of the other 52 sections, and is thus given the same importance as, for instance, Orthodox Philosophy 1, or Orthodox Philosophy 2, or Orthodox Philosophy 3, or the Religion of the Rig Veda, or any one of such outstanding figures as Ram Mohun Roy, Dayanda Sarasvati, Ramakrishna, Vivekananda, Aurobindo, Tagore, Gandhi, and Radhakrishnan. Generally speaking, ideology and ritual are described for their theological and soteriological significance, as though the salvation of the individual soul was central and fundamental, and that Hinduism exists as a religious philosophy that has universal relevance and which only contingently happens to be practised in India. The centrality of Hinduism as an ideology belonging to the social relations of a particular group or set of groups is acknowledged but then side-stepped. Virtually everything that sociology has revealed about Hinduism is ignored in the quest for a soteriological belief system, a World Religion, which transcends any particular social group (4).

A similar criticism can be made of R.C. Zaehner in his Hinduism (OUP 1971). Right at the beginning of his book he says

"It is perfectly possible to be a good Hindu whether one's personal views incline towards monotheism, monism, polytheism or even atheism." (1971:1).

And he says that Hinduism is

"as much a social system as a religion......Indeed until a century ago the acceptance of the caste system was considered by the orthodox to be the sole effective criterion of whether one was or was not a Hindu". (quoted in Brian K. Smith, 1987:37/8) (5)

Zaehner's point leads easily to the notion that to be a 'good Hindu' one must be ritually pure, and ritual purity is related to caste, age, gender and so on. Hinduism is an orthopraxy, not an orthodoxy. Yet the book contains virtually nothing on caste, but is all about the myths and doctrines and philosophies which are presented as though they had no location in a social reality. My point is that, despite this scholar's own observation, he has treated the subject as though it were only marginally about caste and the ritualisation of life, and has packaged Hinduism as an orthodox literate tradition which is a 'religion' ultimately because doctrines about salvation and the human soul are the fundamental criterion for what is included and what is excluded.

My own research in India has been on Buddhism and social change in Maharashtra. This is an analysis of Buddhist institutions in urban and rural Maharashtra. It is also an analysis of the writing and polemics of B. R. Ambedkar, an untouchable who became the first law minister of Independent India in the Nehru government but who then 'converted' to Buddhism.(6)

One aspect of the situation is that almost all the people who followed Ambedkar into Buddhism belong to one untouchable endogamous caste, and the meaning of Buddhism in this context is inextricably entwined with problems of caste ritual, politics and soteriology. These are in fact the basic categories with which I have tried to analyse this Buddhist movement. The notion that I am researching something called religion as distinct from something called politics or caste or more generally 'society' has no substance.

For example, I cannot simply study the soteriology and imagine that I have understood anything significant about the real world of the community known as Buddhists in this part of India. There is an institution called TBMSG based in Pune which has a function in some ways similar to the traditional Sangha in Theravada countries, though in other ways it is also crucially different. It can certainly be described as soteriological. It has a clearly defined soteriological doctrine and it seems legitimate to say that committed members of this organisation behave in certain ways because they believe the doctrine to be true. The soteriological doctrine, as it is framed by TBMSG and its founder the Ven. Sangharakshita, implicates it fully in actions of social responsibility and compassion, and consequently it is involved in important social and educational projects. It fully deserves to be studied and written about. However it would be a travesty to imagine that one could legitimately present this organisation, or even all the soteriological organisations operating in this region, as Buddhism in Maharashtra and cut the analysis off at that point.

To understand at all what is happening in this situation I have proposed an analytical continuum in which politics and soteriology are mixed up and both coexist with ritual, even though ideologically speaking ritual is largely a contradiction of political and soteriological aims. By ritual I here mean a range of ritualised relations between people of different castes, sub-castes, gender, age; between people who live in different parts of the village; between people and mystic powers such as the goddess; and so on. Buddhists, like Muslims, Christians and others, are all located in this ritual network whether they like it or not.

If one looks at any specific situation, for example agricultural peasants in a small village in a remote rural area, middle-class (but still untouchable) academics in a city, urban factory workers, renouncers, one will get different degrees and combinations of ritualistic, political, and soteriological attitudes, understandings, and activities. They all merge into each other in self-implicating ways, or sometimes in quite contradictory ways. But there is nothing which I can usefully identify as 'religion' here, as distinct from anything else. In the sense of religion as values, religion is all of them, and therefore the term has lost any clear analytical meaning. The central explicit values of the movement, such as equality, liberation (moksa), enlightenment, education, freedom from caste oppression, and so on, have different nuances in different contexts.

What I have come to suppose I am studying are Buddhist values, that is values held and institutionalised by specific groups of people who describe themselves as Buddhist. These values include egalitarianism, democracy, education, the rule of law as against ritual hegemony, enlightenment, liberation, peaceful revolution and so on. Many of these basic values require interpretation, for example 'enlightenment' and 'liberation' mean something rather different for urban intellectuals and renouncers, but their meanings are not clearly separated either; the value of education can mean an increase in personal and family dignity, the chance of a good job, a better dowry from a bride, or higher ritual status for Buddhists in general or for a sub-caste in particular.

One can only understand this 'Buddhism' in Maharashtra in the context of a whole range of ritual institutions which affect all Indians, whether they are nominally Buddhist or Muslim or Hindu or Christian. For example, despite a general commitment to egalitarianism, Buddhists are not only treated as untouchables by higher castes, but they sometimes treat other untouchable castes as 'relatively' untouchable to them. Despite a widespread commitment to love marriages as against arranged marriages, virtually all Buddhists practice endogamy and in some areas at least sub-caste endogamy is the norm. Despite an explicit and often-repeated rejection of the worship of traditional deities which is seen as irrational and superstitious, Buddhists are still involved in animal sacrifice to the goddess Mariai but they seem to have cut themselves off to some extent from traditional rights/duties such as scavenging, and also refuse to participate in some Hindu festivals, especially if they are controlled and organised by the higher castes such as Maratha. But this refusal is highly problematic since it involves upsetting power relations as well as ritual relations, and the Buddhists have suffered several pogroms in this region. There is also a general Human Rights problem connected to indebtedness which leads to virtual slavery, known as bonded labour. This is not a problem for Buddhists only but it is certainly an important feature of their situation. To discuss the 'religion' of this region without mentioning the problem of bonded labour, the practice of dowry, and the related issues of prestige and status in relation to gender and caste, would seem to me to require theoretical explanation and justification.

Again, to study the Buddhists without putting their value for education or political egalitarianism, or their actual practice of untouchability, near the centre of one's analysis because of some prior commitment to a vaguely conceived religious studies would be to move away, rather than closer to, people's actual lives. It would be to substitute an ideological (I would say disguised theological) commitment for understanding.

In a way these points seem obvious, and religion departments are full of scholars who are well aware of these kinds of issues. And yet the idea of religion as some substantive aspect of human existence, or some distinctive institution which justifies having separate departments and a special publisher's list continues to ghost even books of considerable theoretical sophistication. But the examples which I have given so far, of American Indian culture and of India suggest that 'religion' is a pretty useless category. One cannot pick out some of these values and institutions and say they are religious and designate the others as non-religious. It is a phoney problem generated by a distorting concept.

Let me now turn to the case of Japan. The so-called religions of Japan are frequently listed in books as Shinto, Buddhism, Confucianism, Christianity, the New Religions and the New New Religions. An example of this is Japanese Religion: A Survey by the Agency for Cultural Affairs (ed. Hori, 1972, reprinted 1990.) (7) However, each of the authors who contribute articles on the so-called different religions suggest that their own field provides the fundamental context of values for understanding Japanese culture and society. The article on Shinto is unambiguous:

"Shinto in the most comprehensive sense of the term is simply the basic value orientation of the Japanese people in the various forms it has taken and the developments it has experienced in the course of Japan's history...". (1972;1990:29).

But then Mitake Hitoshi, in his article on Folk religion, says:

"It is within the frame of reference provided by Folk religion that the organised religions have made their way into Japanese society. Only as they accommodated themselves to folk religion and its implicit norms did the institutional religions find acceptance and begin to exercise influence on the people in their daily lives. Any attempt to understand the role of Shinto, Buddhism, and other religions in the lives of the Japanese people will of necessity have to come to terms with folk religion." (1972:122).

Mituo Tomikura, writing about Confucianism, says:

"Confucianism has exercised an enormous influence on Japanese modes of perception and behaviour through its concrete connection with concrete problems of human relations in daily life, Confucianism was brought into touch with the traditional religions at the point of their practical foundation in society and had an important influence on the shape they took." (1972:105).

Tamaru Noriyoshi's account of Japanese Buddhism does not claim it as a fundamental element in the development of Japanese culture. Instead, he emphasises the way Buddhism was forced to conform to the general contours of the Japanese value system, for instance how its distinctive universalistic ethical and transcendental elements were eliminated and transmuted into ritual and hierarchy in the ideological fusion with the Shinto and Confucian elements. (1972, 1990:47; passim).

Probably, an argument about which of these elements is more basic in the formation of Japanese culture is unfruitful. The essential point, as Professor Shigeru Matsumoto points out in his introduction, is that we are dealing with one cultural or ideological complex:

"The newly introduced traditions did not uproot the indigenous but were invariably assimilated into a kind of homogenous tradition which itself might be called the 'Japanese Religion'." (1972: 39-40).

The point I am making here is that the elusive 'religion' substances weave their way in and out of the text even at the very moment that the authors are saying that the substances do not really exist as separate entities but are merely ancient sources of values and institutions which form a homogeneous tradition. These writers are all in their own way identifying a system of ritual prescription tied to status, hierarchy, particularism, lack of doctrine, lack of universalism in ethics and philosophy, symbolic sociological awareness and so on. The use of the concepts of 'religions' such as Shintoism, Buddhism etc. determines the structure of the books organisation, and yet the scholars' analyses show that it is actually distracting us from the real situation, which is the dominance of various core values such as deference and hierarchy which are reproduced through the agency of many different institutions, some involving reciprocal exchanges with mystical powers.

This same point is, I think, further clarified by considering another recently published book Religion and Society in Japan (ed. Mullins, Shimazono, Swanson, 1993). This book has some excellent features. However, the different authors in this book use the word 'religion' in relation to just about anything, for example soteriological doctrines, supernatural technology, belief in angry ghosts, ancestor worship, buying a Christmas cake, using amulets and charms, civil religion, basic values and so on. On the other hand, nobody actually discusses what the object of study is, and why some things have been included and others excluded.

Paul Swanson, in his introduction to Part 1 (Japanese Religiosity), picks out Kuroda Toshio's point, in one of the more difficult articles, that

"...the simplistic understanding of Shinto and Buddhism as two independent religions is misleading at best for much of Japanese history, and the same is just as true for contemporary Japanese society". (p4)

Swanson links this to Miyake Hitoshi's paper on Shugendo which he suggests makes much the same point. I think Swanson has in mind Miyake's argument that Shugendo supernaturalism was an integral part of the village religious economy during the Tokugawa era, and has since been absorbed into many of the New Religions.

I will argue that Swanson's (and Kuroda's) point is too significant not to be taken further. For if Buddhism, Shinto, and Shugendo are not fundamentally independent religions then the obvious implication is that they cannot be adequately analysed as such. An analysis of Shugendo, for example, or Mahikari, or Soto Zen Buddhism, would need to be placed in a wider and deeper context of analysis.

To take one example: Shugendo rituals, Mahikari rituals, rituals performed at a factory and the Yasukuni Shrine rituals are all treated separately in different papers with no cross-linking, as though they existed in a vacuum. Yet they are all linked in fact by the prevalent belief that angry, malevolent spirits who died a 'bad death' (goryo shinko) are a threat to the safety not only of individuals but of institutions including the family, the factory/corporation, and the nation itself.

My suggestion is that an analysis of the symbolic or structural meaning of these dangerous spirits could provide us with a more comprehensive picture of what the real subject of study is. Concepts of hierarchy, purity and pollution are especially illuminating in this and other contexts, for not only do they connect the phenomena with each other, but they also place them in the wider semantic context of dominant values which other scholars such as Chie Nakane have written about. And even if my suggestion turns out to be wrong, or only half-right, it at least seems more fruitful, and more likely to generate insights and ideas, then the vacuum-packed presentation of these phenomena as instances of some kind of supernatural technology unconnected to anything else in the culture in which they are embedded.

A concept of religion defined by doctrines concerning the supernatural, salvation, life after death and separate, clearly demarcated institutions does not provide a satisfactory analysis in the case of Japan. Scholars themselves use the word 'Religion' and 'Religions' in so many different, and usually imprecise ways, that it is unclear what the real subject of study is.

I argue that the adoption of a concept of cultural studies, based on the notion that values and their institutionalization are the real field of study, produces a more fruitful and less distorted analysis. Connections which are marginalized in the study of 'religion' become suddenly central when we analyse Japanese society as a system of values dominating a wide range of institutions and expressed in the ritualization of social relations. Dealings with mystical powers are an aspect of this picture.

An important aspect of my proposal is that temples, shrines, graveyards, and household shrines, where transactions with the supernatural usually take place, may be best understood as acts of ritual reciprocity undertaken to protect the purity and orderliness of the Japanese world from the pollution and disruption of foreign or ambiguous elements, or put more simply to protect the safe and familiar inside (uchi) from the wild and dangerous outside (soto).

I suggest, following Chie Nakane, that one important (probably fundamental) value is the senior/junior relationship (oyabun/kobun; sempai/kohai) a principle of hierarchy which carries the notion that the relationship (the whole) is fundamental, and not the individuals who relate. This senior/junior relationship is one of mutual dependency, obligation and obedience. It emerges in different ways in a variety of different social relations, and acts as a nucleus for a whole range of other important values, such as reciprocal obligation (giri/on), sincerity (makoto), harmony (wa), deference and so on.

The ideological priority of the hierarchical relationship over the individual was also discovered by Edwards in his work on Japanese marriage ceremonies and usefully formulated as the incomplete individual (8). Mature personhood is achieved in Japan when one recognises that one is incomplete, that one is dependent on other. The notion of the self-sufficient individual is seen as childish and selfish. Furthermore, this central cluster of values, being reproduced at different levels and in different contexts, are associated strongly with Japaneseness, which some writers feel is the 'true' religion or object of veneration (Nihonkyo). However, Japaneseness, 'being Japanese', while being a racial concept of birth, blood and language, as well as location or national territory, is also a ritual concept, for it means being located in a sacred land in a hierarchical network of relationships which provides each person with a meaningful identity, and which is celebrated in a multitude of ritual contexts.

This way of looking at Japanese values and their ritualisation suggests that the temples and shrines are not the only or even the main institutions where the dominant values are passed on from generation to generation. Edwards argued that the katei (family), though often a nuclear family household in the sense of physical separation from wider kinship networks, nevertheless embodies some of the same structural values as the traditional ie (household). And Nakane Chie, Yoshisa Abe and Kunio Yanagawa have all argued that the modern corporation has taken the place of the traditional ie or household in important ways. The great corporations of Japan are inevitably important institutions in the reproduction of the social order. Lewis's data on spirit rites in a Japanese factory, though presented in my view naively as merely a supernatural technology for dealing with danger from fires, can be made more sense of as at least in part symbolic of the ethos of the corporation. And Swyngedouw has suggested that Zen meditation techniques are used by some companies "to deepen human relationships and to teach proper etiquette and strict adherence to the company rules". If Swyngedouw is correct, it would be difficult not to see Zen meditation techniques as having some significant (and traditional) relationship to social control and the reproduction of the Japanese social order. This in turn would put a semantic spin on 'soteriology' which Swyngedouw has put his finger on because he thinks as a sociologist (even though he is also a Catholic priest) who lives in Japan and confronts Japanese reality everyday.

The Soto Zen sect, as described by Reader in a different article in the same book, is more concerned with the reproduction of hierarchical values in the context of the family, and the maintenance of the ancestral spirits (which are fundamentally part of the household), than with the salvation of the individual. And I myself have argued in my article 'Japanese Religion as Ritual Order' (Religion 1992) that the school system is a fundamentally important institution for reproducing the dominant values of the Japanese sociocultural order. To study shrines, temples and sects but not to study the schools seems to me to be theoretically or methodologically unjustifiable. It is the somewhat analogous case to claiming to study a religion called Hinduism, packaged around gods, liberation and righteousness, while ignoring caste, gender, untouchability, and politics.

Attempts to redefine the concept of religion to take the broader context into account and thus to save it from extinction tend to dissolve it into ideology or values generally. In particular I have in mind arguments based on the idea of family resemblances (9). On this argument, there is no essential property of 'a religion' or 'religion' in general; religions, for example, do not need to be defined in terms of belief in the supernatural; there are a series of over-lapping features which no specific instance of religion need necessarily have. I cannot argue this point in detail here (10). I merely want to point out that the actual uses of the word religion by scholars in their publications are so varied that the family is truly gigantic and amorphous. This line of argument merely moves the notion of religion towards dissolution and, I believe, implicitly if unwillingly supports my contention that the real object of our study is values as they are institutionalised by specific societies.

If one actually looks at what religion scholars are studying, the most effective common denominator of the vast range of historical, textual and sociological data is the concept of values institutionalised by specific societies or groups of people in oral traditions, texts, gift exchange, marriage, dealings with mystical powers, conceptions of property and inheritance, and so on. By and large 'religion' scholars who are not doing straightforward theology, or the more covert variety called phenomenology, are in fact studying values and their institutionalisation in different cultures by observation, ethnography, textual and historical studies, all of which are essentially hermeneutical, depending on  very sensitive contextual analyses and interpretations, with a self-critical eye on the match between emic and etic concepts and the ideological loading of their own tools of analysis.

Therefore I am arguing for a clear and open distinction between religious studies as a branch of theology and religious studies as the study of values as they are institutionalised by specific social groups in their myths, rites, texts, architecture, political institutions, conceptions of legitimate power, and so forth. As a contribution to this reconceptualisation, my argument is that, whether or not 'supernatural' or 'transcendent' entities are involved in the study of values, and in what sense, and whether or not some notion of personal salvation is involved or not, is itself a problem of interpretation, for the meaning of these terms is not self-evident. There is nothing a priori for or against their inclusion in the field of study. The crucial thing is to shift the analytical weight clearly and openly from mystical powers to human institutions, from gods to valuesWhy has 'religion' persisted as the centre of a whole academic field and publishing genre if it is really an illusion? How can an illusion have power?

This is a complex historical subject (11)and I can only allude to it here. But one relevant aspect has been the colonial context, the relation of coloniser to colonised, and within the colonised the relation between a dominant elite and its own dominated people. I suggest one needs to think about 1) the invention of 'religions' not only by imperial civil servants and missionaries but also by local elites such as neo-Vedanta in India, Protestant Buddhism in Sri Lanka, the religions of Japan since the beginning of the Meiji; 2) the growth of ecumenical dialogue between the dominant Christian religion and the religions of the elites of the colonised countries. Theosophy was one of the ways in which this ecumenical dialogue got started.

There is a fairly simple idea underlying the complexities of theosophy, and that same basic idea comes up in various forms in Eliade, in Huxley and the Perennial Philosophy tradition, in Wilfred Cantwell Smith, and in influential philosophical theologians such as John Hick. It is the notion that there is one transcendent reality and that different religions are partial and incomplete manifestations of it. In the work of John Hick or Don Cupitt, for example, this idea is an openly philosophical and theological issue, and is surely connected by others with improving communications between the so-called world faith communities. But in the work of Otto and Eliade the essentially theological idea was disguised (not very well) behind phenomenology or the science of religion. Phenomenology has helped an essentially theological agenda get by as a 'neutral' and 'objective' descriptive science.

It is this disguised connection with a theological agenda which I believe helps us to understand the persistence of the religion genre, its tenacity and the strength of its institutional basis. Even though many scholars will frequently, at moments of critical reflection, explicitly reject that their work is anchored to the notion of the supernatural, to doctrines about the transcendent, or to some sui generis reality denoted by the word 'Religion', its institutional history nevertheless carries such powerful theological resonances, disguised by the cloak of phenomenology as an objective human science akin to sociology but distinct from sociology, that it tends to draw even these scholars unwillingly into its orbit, thus creating distortions in the field of studies and holding us back from a simple and straightforward reconceptualisation.

I suggest that many scholars who happen to be employed in Religion departments and who publish within the religion genre have no personal theological agenda and are in reality studying values in specific sociocultural situations, either as historians, or linguists, or sociologists, or anthropologists, or whatever. The false distinction between religious and non-religious institutions and values is  not a working concern of many scholars. Asking whether any particular value such as deference should be categorised as religious (because it is directed towards a dead ancestor) rather than non-religious (because it is directed towards the sensei) would seem - if it were pointed out to them - to be missing the humanly central issue. Analysis and interpretation of the value of deference itself, as it is understood and institutionalised by a specific group of people, would seem to be the key issue, and is in fact treated as such by many scholars working ostensibly within a field called 'religion'. And yet at the same time in books about 'religion', 'religions' and 'world religions', the tendency will be to produce the material on the ancestor worship but not on the relations with the sensei. Guided by some vaguely formulated a priori assumption about religion, the focus on deference and those disparate institutions which it fundamentally connects is missed.

My argument then is that the only thing the continued use of the word religion does in a scholarly context is damage, since it creates the illusion that something substantive is being picked out and analysed. What is actually happening, in my view, is that analysis is frequently being impoverished by being cut off at the wrong places under the distorting influence of a partly unconscious idea. What I want is for the study of institutionalised values and their relation to power to be clearly and consciously separated from disguised theology in the field which is now covered by 'religion'.

Notes:

(1)In this paper I give a summary of the argument of an almost completed book, for which I have not yet found a publisher.

(2)One recent book worth considering in this context is Fred Inglis, 1993 Cultural Studies, (Basil Blackwell).

(3)See:Stewart Sutherland et al. eds. , 1988 The World's Religions, Routledge:873.

(4)I have discussed this and the following book, and the case of Hinduism more generally, in my article 'Hinduism and the World Religion fallacy', Religion 20:2 (1990) pp. 101-18.

(5)See for details: Brian K. Smith, 1987 'Exorcising the Transcendent: Strategies for defining Hinduism and Religion' (History of Religions)

(6)I have published on Ambedkar and the Buddhist movement in Japanese and Indian journals. My most recent paper is 'Village Buddhism in Maharashtra' in Indian Journal of Buddhist Studies (Sarnath) 1994, Vol. V, Pt.2.

(7)I have discussed this book and these Japanese issues in general in my article 'Japanese Religion as Ritual Order' in Religion, (1993) 23, pp315-341.

(8)See: W. Edwards, 1989, Modern Japan through its Weddings: Gender, Person, and Society in Ritual Portrayal, StanfordUniversity Press.

(9)Two people who have adopted this strategy are Ninian Smart 1973The Science of Religion and the Sociology of Knowledge and Peter Byrne in Stewart Sutherland (ed), 1989, The World's Religions (Routledge). There is also the anthropologist Benson Saler who uses this formulation in his book Conceptualizing Religion: Immanent Anthropologists, Transcendent Natives and Unbounded Categories. However, I do not have his text yet.

(10)See my article 'Hinduism and the World Religion Fallacy' (see note 4) for a more detailed critique.

(11)Wilfred Cantwell Smith, despite his theological agenda, has given an interesting account of the invention of a modern concept of religion in his book, 1962. The Meaning and End of Religion .