PART TWO Religion, Myth and Ritual
Stephan Feuchtwang
A CHINESE RELIGION EXISTS
In his latest writings, Maurice Freedman recommended and himself worked with the assumption of a single Chinese religion. Religion, according to him, consists of heightened expressions of principles of social organisation. This notion will be expounded, but I will also argue that Freedman, with great rigour, pressed the notion to the borders of its capacity, at which point it begins to break down in interesting ways.
Consider, to start with, this dialogue composed from the pages of religion and Kitml in Chinese Society (edited by Arthur P. Wolf, 1974):
Freedman: 'There is some order—of a kind that should allow us (if we take the trouble) to trace ruling principles of ideas across a vast field of apparently heterogeneous beliefs, and ruling principles of form and organisation in an equally enormous terrain of action and association. Ideas and forms need not be uniform to be common; they may be reflections, perhaps misshaped reflections, or idiomatic translations of one another....' (p. 21)
Wolf: 'How do we distinguish full and accurate translations from mere metaphors?' (p. 2)
Freedman: [We assume a country, and in this case it is] 'a country of China's extent and political cohesion...one might predict from first principles that a society so differentiated...allowed religious similarity to be expressed as though it were religious difference.' (p. 38)
Wolf (citing Robert J. Smith): 'It is equally likely that this society may instead have treated religious differences as though they were religious similarities.' (p. 3)
140    Stephan Feuchtwang
What is Freedman's 'community of religion' (ibid.: 40)? It is described as occupying a vast field of apparently heterogeneous beliefs in which it is possible to discern ruling principles or ideas and an equally enormous terrain of actions and associations in which it is also possible to discover ruling principles of form and organisation. But the differences between ideas and between forms may be thought of as translations of one another, like the idioms of a single language. For Wolf, on the other hand, it is the heterogeneity that stands out: not one language, perhaps, but many— though all certainly members of the same family of languages. For despite their different points of view, Freedman and Wolf do apparently share a basis of agreement. Both accept that there is a framework ('a country' or 'a society*) within which both the coherence and the differentiation exist. That the framework encloses great social complexity together with political centralization and cultural identification is not in question. Issue is taken only on the place of religion in the coherence of this terrain.
That religious expression helps link the different points within the framework in a process similar to that of translation is not in question either, but mere linkage does not exhaust the matter, since both writers consider that religious expression carries an intrinsic quality of its own that necessarily expresses something 'religious'—a particular kind and quality of belief implying profound identity. In Wolf's traditional understanding, this particular kind of belief is 'concepts of the supernatural' and its particular quality is authenticity. 'Take care', he writes, 'to distinguish statements that reveal belief from those that use belief to express social distance' (ibid.: 3). Accordingly, the controversy is about the placing of profound identity: does the framework enclose a single identity that appears in different forms like the different versions of a single text, or does it instead surround a multiplicity of different identities each to be understood only in its own context? These questions are bound up in a simpler one: namely, what is the China which contains one or many religions in the first place?
Freedman made the most serious and sustained attempt to add a Chinese dimension to the theoretical preoccupations of the British social anthropology of the 1960s and 1970s. His attempt always included the rejection of village studies. The little ethnic units of islands or villages are only the most expedient and, in their endorsement of parochial horizons, deceptive extrapolation of China for social anthropological thought. There are other ways of turning Chinese ethnography towards anthropological problems. Amongst them are Freedman's 'lineage systems' (1958 and 1966) and 'marriage rites'- (1967a and 1970a), which can rank as contributions definitely aimed at constructing new versions of the anthropological notions of agnatic corporate groups and of marriage ritual. They have proved to be a productive legacy, for they are cases to which much has been added and is being added. But they remain case-studies of Chinese
A Chinese Religion Exists
institutions, unabsorbed into the general body of social anthropological theory.
More ambitious would be to pit China itself as the conception of a whole social order against other conceptions or preconceptions, in the manner of the India whose uniqueness in Louis Dumont's work provokes Homo Hierarchicus into existence and a new version of the uniqueness of a rationalist and egalitarian West (e.g. 1966). Some such ambition might have been behind Freedman's reassessment of the work on Chinese religion by two sinologists, de Groot and Granet. He was promoting and reassessing them as part of his new assertion of the existence of a religion of all China and the close attention he intended to pay to Western perspectives on 'religion'.
The notes1 he wrote for the first of three lectures in 1974 to introduce de Groot and Granet on Chinese religion, under the general title 'The Western Perception of Chinese Religion',2 state:
My chief aim is to illustrate (from Chinese data) difficulties attending description and analysis of exotic systems of ritual and 'belief, (a) We have a notion of what our religion is and we are tempted to measure other people's religion by that rough standard, (b) We change our views of what analysis of religion ought to be (under inspiration from developments in social sciences), and our perception of exotic religions changes accordingly. The chief problem under (a) is that we may too easily assume that religion is a universal category in the sense in which we mean it in our civilisation in the nineteenth and twentieth centuries, i.e.: as something separable and delimited, and with a name. [AJlthough in very modern times the Chinese have produced a term tsung-chiao [%ong-Jiao ^ffe] (coined in fact by Japanese) to correspond to the Western category, they do not in fact think in the fashion implied by our category. In traditional China you could ask a man about his chiao \jiao fjf] (his doctrine, his school); you could question him about his rituals and festivals. But there is no way of getting him to conceive of a separable area of cultural life corresponding to our word religion. It follows that the exercise calls ideally for a 'hermeneutics' of our prime categories before starting that of categories of exotic religions. The prime categories are: religion, magic, divinity, worship, sacrifice.... As in all anthropological endeavour, the exercise should end by our having a clearer notion of our own religion. The study of the exotic is one means of self-study.
This, then, was the programme, in the fullest statement I know, upon which the work of Freedman's last years was set. The aim was to question the concept 'religion' by placing it and its usage historically and biograph-ically. But the result envisaged was not to de-universalize the concept,
1. I should like to offer my warmest thanks to Judith Freedman for allowing me to use these and others among Freedman's papers.
2. These were lectures which he would have given in 197; had he lived, repeating with a few modifications and additions what was already published by him on de Groot (Freedman 1974, 1975 a) and on Granet (Freedman 1974, 1975 c).
Stephan Feuchtwang
rather to change it in order to approach a truer universality, and, for a start, to test its correspondence to its Chinese reality.
The first, formidable problem faced in that test is, as the above excerpt states, that religion does not exist in China as something separable. Beyond that problem looms the risk that 'religion' might not be salvaged at all. The empirical test is to find the distinctions entailed by the definition of 'religion' in each ethnography of religion.
Ever since the reports of the first missionaries and merchants were taken up by Enlightenment philosophies, the difficulties posed by a Chinese 'religion' have been an embarrassing disjunction of categories thought to be necessary to each other, and the reverse conjunction of the unexpected. Chinese religion has provided an evident ethic and the development of universal doctrines without the support of a Supreme Being; an evident state and polity without justifying mysteries; evident mystery, prophecy and ecstasy without speculative ethics or anti-ethics; religions in the sense of institutionalised and centralised orders but no central corpus of doctrine, no finality, no community of 'conscience'.3 Can some sense of religion be retrieved from these apparent discrepancies? Can social anthropology be the saviour which can restore a universal sense to the concept of religion?
Since the Enlightenment, China has stood as one or other of a pair of figures. The figure of enlightened administration and secular ethics for European, eighteenth-century rationalism is one. The other, the figure of monstrous materialism and benighted superstition and despotism, was the China of nineteenth-century free-thinkers and Protestants promoting 'conscience' and protest. A commentary on the first, Enlightenment figure would remark that the administration of the wealthiest empire in the world did after all include religion, but that it was an administered religion: the non-rational, the non-prosaic, the transcendental were not public affairs. This remark would be re-written for the second, free-thinking figure. In this version, China excludes religion, and thereby debases the rational into the immediate, the symbolic into the prosaic, the transcendental into the material. Superstition is ignorance and therefore not religion, it would say.
The changing emphasis between earlier and later stages of the work of the late nineteenth-century sinologist de Groot repeats each of these figures through the grid of an evolutionary and subsequently a degener-ationist thought that was of his own times. Freedman's lecture notes on de Groot summarise his movement away from the folklore study of a fellow civilisation, a study which Freedman dubbed 'China Coast Aryanism' and obviously preferred to the later de Groot. From an approving description
3. 'Conscience' translates the single French term covering both consciousness and moral values.
A Chinese Keligion Exists
of moral education provided without religious sanction and of religions harmonised and in their place within a regime of tolerance, 'the Chinese (in de Groot's work) have begun to move from civilisation to semi-barbarism, from folklore to ethnography; that transition is accompanied by a movement from a total to a partial perspective in the sense that it now seems evident that there is a proper form of Chinese religion enshrined by a superior class of society'.4 And the proper form is a stultifying conservatism; to the later de Groot 'Chinese religion was a total system, and it was a total system in error, which in the end would be overturned only by modern science' (original emphasis). Finally, Freedman claims in de Groot's commercial-cum-diplomatic dealings with the Chinese, 'the residue of the eighteenth-century view of China incorporated into the Aryanism of the 1870s was annulled by the tougher attitude towards China displayed by Europe as the century moved towards its close. The anti-Catholic Deist young de Groot was transformed into the pro-missionary de Groot.' Let us see how Freedman's social anthropology could have pursued a 'Chinese' critique of 'religion' from this point.
Keligion
Some anthropologists declare that [the word religion] cannot be precisely defined any more than some other terms of art in the subject, such as 'kinship' and 'culture'.
Freedman 1979
That 'religion' is a term of art does not mean that it carries no weight and gives no direction. On the contrary, its creative direction can be all the more forceful by being unspecified; 'religion' is one of the standards of social anthropology's unquestioned assumptions. Something of its force in Freedman's work may be gauged by first asking what there was to admire in the stance of the early de Groot. It was a kind of humanism which is not social anthropology and not what Freedman adopted, but which still gave results of the kind which he thought was wanting. Let me inquire further by adding another name to Freedman's pair of ethno-sinologists, Granet and de Groot:
Of the four writers listed by de Groot in his journal as the chief examples of his reading on religion while he was a student, only two are likely to hit the mark with a modem audience: B.F. Strauss and Ernest Renan (1863). (Freedman, 'Western Perception...', 2nd Lecture, p. 3)
4. This and the following quotations are taken from Freedman's lecture notes, 'The Western Petception of Chinese Religion: J. j.M. de Groot', 1974.
144    Stephen Feuchtwang
Renan (1863) stands apart from the materialism of Strauss and from the various evolutionist theories of the history of religion and its surviving traces (for the preservation of which Chinese religion was particularly valued by de Groot), because he stands as the bearer of what was necessary to the others—namely, the elucidation of the notion of a common origin of all religions. As Freedman summarizes this notion (the first of the propositions which he said were implicit in what de Groot wrote), it was that 'there is one ultimate human condition from which all forms of religion spring' (ibid.: 6). Spinoza had already written:
The greater our knowledge of natural phenomena, the more perfect is our knowledge of the essence of God (which is the cause of all things). So, then, our highest good not only depends on the knowledge of God, but wholly consists therein; and it further follows that man is perfect or the reverse in proportion to the nature and perfection of the object of his special desire; hence the most perfect and the chief sharer in the highest blessedness is he who prizes above all else, and takes special delight in, the intellectual knowledge of God, the most perfect Being. (1951: 59-60)
Renan and others, in turn, substituted for Spinoza's assured object of an ultimate explanation and a singular Being in all extent and in all activity, a human impulse of striving for ultimate knowledge itself; a further subjectification of the universe, so to speak. For them, the history of religions can be written as the revealed text of human 'conscience'.
The presence of some of those qualities which had previously been held to constitute the presence and the knowledge of a Supreme Being (wholeness, singularity, infinity) now became the properties of humanity and its 'conscience'. Other qualities (perfection, or ideal reality, ultimate completion and knowledge) were transposed into the objects of conscience's particular movement, its desire. Religion was no longer the body of rules of devotion binding a community to obey Divine Law, nor was it the life of seeking knowledge of God through natural reason. Instead, religion now became that which most directly (if never completely) represents, expresses and records the most internal region of human life: shared subjectivity. To go further and locate the plurality of shared subjectivities which circumscribe the individuals of humanity is the work of social history, ethnology and sociology. But in them, 'religion' retains its position as the most direct expression of an inner, but not individual, nature. Granet, for instance, was said by his director of studies to have found texts 'containing the formulae of rites which have remained the basis of Chinese society. Such rituals offer the interest that they inform us on the religious aspect of phenomena, which aspect constitutes its [sic] inmost nature' (Freedman 1975 c: 9).
A student of ethnology and all later empirical anthropology learns to use the neo-theological distinctions of interiority such as that between
A Chinese Keligion Exists
essence and expression. But the seriousness of the empirical undertaking lies in the methodological imperative to show that any ideas, including the very distinctions by which the internality of ideas is inferred, exist in the subject-matter of analysis. The result, if successful, would be to render that subject-matter transparent, both an object and an example of the analysis and interpretation themselves, their objective existence, their self-evidence.
The self-evidence of shared subjectivity could then test some of the standard philosophical categories of subjectivity. Indeed, descriptions of religion do both entail and seek certain distinctions in the categorization of subjectivity. Among these distinctions are such pairs as those connected with knowledge (rational and experimental versus transcendental and non-empirical), calculation (rational versus tradition-bound), representation (prosaic versus poetic), purpose (conscious versus unconscious) and ends (immediate versus ultimate). Freedman's own summary description of religion as one of social and cultural anthropology's 'main fields of study' is a case in point: religion is a 'mode of experience and belief and 'a way, however difficult of access, of knowing and apprehension', but a mode or a way to be distinguished from the conscious intention and behaviour registered in politics and economics (Freedman 1979: section C6, opening paragraphs).
Whatever distinctions are made, subjectivity is ultimately single, a point of view or experience. On the other hand, subjectivity is necessarily an assumption, since it is described as an interiority of agency at a point of ultimate inaccessibility. Its assumption leaves only the questions, where is that point, or how is that unity composed and how fixed?
Anthropology could then be described as the universalization of this assumption in empirical study. Subjectivity is the identifying quality by which humanity and the irreducibility of its individuals are to be defined. The units, conditions and variety of shared subjectivity are the descriptive problems of anthropology. But problems they must be, since description is, following the assumption, always from the 'outside'. The terms of . anthropology's art fly apart in three directions, to an 'outside' secure only in the designation of »o«-human conditions or origins, to an inside within the art of anthropology itself, and to the other inside within the object(s) described. Any ethnography, and, what is more, any of the terms which by a universal applicability would make ethnography possible, undergoes the test of this flight.
Consider once more Freedman's advocacy of de Groot's descriptions of Chinese religion. Though not advocacy of a methodology in every respect, it is advocacy of a methodological assumption that the religion to be described, however 'religion' must be redefined in the process, is commensurate with 'its' ethnic unit, China-as-a-whole.
This approach goes some way toward helping to solve the problem posed by religion not having a separate existence in China. If religion is
Stephan Feuchtwang
defined as privileged points of access to shared subjectivity, it can be found diffused in its social context. Even though the unity of this shared subjectivity does not or does not usually find separate (religious) expression, nevertheless the mere existence of a basic order in social life itself implies that there is such a unity. Thus, if we are to locate the religion, all that is required to validate the claim that the multiplicity of rituals and beliefs do in fact constitute a unity is to assert the existence of a social identity. However, difficulties are bound to remain if, instead of simply accepting the constitution of the unity of context, i.e. 'China', as a working assumption ('we are all working on religion in the same field'), we insist instead upon demonstrating its nature. And Freedman did demand this of his own work. When the unity, which Freedman referred to as 'Chineseness', is, as we shall see, defined as a complexity of social organization with no key principle, that complexity threatens to replace any identity of which it is the context and which it composes.
The religious totality of China is, according to an expression of Granet which Freedman found particularly striking, 'neither God nor Law'. It is constituted by 'a few basic principles of thought and order'. This much is editorially drawn to our attention more than once in the introduction and notes that Freedman wrote for his translation of Granet's Religion of the Chinese People (Freedman 1975 c).5 Such statements are both methodological recommendations and their empirical justifications, adducing Chinese representations of totality. Granet and de Groot, while criticized for their (each quite different) sinological biases, are at the same time recommended rather than the 'parochial vision' of subsequent investigators of religion in Chinese society. Social anthropology had so far failed to provide the necessary correctives to de Groot's and Granet's total perspectives on Chinese religion (ibid.). To do this, a new sinology would be needed.
To this end, Freedman proposed that a complexity of cross-cutting juridical principles—neither God nor Law, but thought and order—were represented in certain rituals and beliefs. But more than this was required. The importance which, in a new-found advocacy, Freedman (1974) attributed to C.K. Yang's sociology of Chinese religion (e.g. 1961) was in its identification of some of the unifying principles and totalistic beliefs as they appeared diffused and refracted through social complexity. They represent the 'Chineseness' of the complexity, however conflicting and differentiated its internal relations are. Of Freedman's own investigations, geomancy (feng-shui), the Chinese art of divining sites of buildings and tombs, stood out as a special example of such a representation. It functioned, according to him, not as a cohesive seal but as a quasi-linguistic medium of transposition between neo-Confucian philosophical
5.   For example, p. 27 and editorial notes 28, 83 and 144.
A Chinese Religion Exists 147
writings, the respectable divination of sites, and ordinary ritual and belief, or between roles in ruling, middle and commoner classes.
Suppose, however, that the analogy with language were taken beyond notions of reflection, expression, representation and translation between the subjectivities represented. To do this would be to run the risks similar to those entailed in the rigorous description of the complexity of social organization by means of juridical analogies. Just as the latter threaten to depart from the necessity of any assumed unity and become mere rules of order, so the former can float off the unity as mere surfaces of expression itself, when the signs and rituals are studied as distinct materials of a linguistic or semiological order. However, these risks are lessened by analysis of 'Chineseness' in which the unity of the juridical order is represented by the symbolic order, and the symbolic order is seen as a representation of the principles of the juridical order in the special dimension of 'religion'. Granet is an important support here.
Granet was more important to Freedman than was de Groot. Partly this is because Granet was trained as a sociologist, and in a school of thought adopted and adapted in British social anthropology. 'In his three years at the Fondation Thiers', wrote Freedman (1975c: 10), Granet 'put Durkheimian sociology to work in the analysis of ancient Chinese society and fastened upon the family and ritual as, for him, the central problems to be tackled'. And again: 'in a sense...religion for Granet was at the intersection of the mythique and the juridique' (ibid.). How easily could this be taken as self-description? Take away the term 'ancient' from Granet's title (La Chine antique), and the problems could certainly be said to be Freedman's own. But in his work are they 'Durkheimian' problems? Nowhere does he, like Granet, place reliance on such notions as the effervescence of mass activity at places and occasions demarcated for the unconscious social purpose of solidarity, and thus held to be sacred. Freedman makes no use of any theory of sentiments. Nor has he any recourse to a universal history of religion. Yet he does, of course, accept in a general sense that religion is a social reality. More, it is the form, if not the formation, of 'categories of thought: categories of class, force, time, space, and so on, and above all, of totality' (ibid.: 19). But totality for Freedman is not what he took it to be for Granet, a reference to original, ancient, historical representations of a social whole. These are but a part of a subject-matter whose inherent modelling is to be reconstructed. Freedman's totality is structural. But in more than one way, this structure too is different from the structuralism attributable to Granet. First, it is not premised on an essential reality of solidary representation. Secondly, it is more juridical than a structure of thought. Freedman's totality is representational, but his notion of representation receives little of the philological and later semiological charges that Maussian anthropology has carried. Rather, it carries the idea of an expression of inner jural
Stephan Feucht&ang
reality, a representation of 'ruling principles' of organisation.
But Granet was, after all, immediately important to Freedman for vouchsafing the discovery of a unifying representation, a Chinese expression of totality, a Chinese humanity, interpreted though it was in Freedman's own fashion. I shall eventually arrive at what this was. It is necessary and possible first to substantiate what was the new, social anthropological, sinology of 'Chineseness' upon which Freedman was intent. This can be done by working through all of Freedman's earlier attempts at describing Chinese ritual institutions. To these I shall add consideration of what he wrote on ethnicity. In this way I hope to trace a way back from the latest to the earliest of Freedman's writings on ritual and religion and then return once more to the latest. In doing this, we. shall find ourselves led a number of moves away from considering Freedman's notion of 'religion' and towards the less obviously philosophical ground of his descriptions of 'ritual'. It is not that there are two Freedmans, early and late, but there is a shift. Late Freedman uses the earlier descriptions of Chineseness increasingly deliberately, taking Chinese-ness and the entire Chinese social order less as a working assumption for the description of certain changes than as a task and a challenge.
Chineseness
Ideas and forms need not be uniform to be common; they may be reflections, perhaps misshapen reflections, or idiomatic translations of one another, as in their transmission back and forth between social strata, between sect and 'church', between 'church* and 'church', between text and living language, between the cultivated and the popular. Their Chineseness lies in a basic stock upon which complex social and intellectual life works and elaborates variety.
Freedman 1974:21
The new sinology would encompass knowledge of fully literate China with knowledge of the rest of the field 'China'. It would find the sectarian or elite views of the peasant masses' religions (or superstitions), as positions in a 'total religious field' (ibid.: 37). But Freedman opposes the local study which gives a 'peasants'-eye view of China...that inverts but otherwise reproduces the distortion of the Confucian-eye view' (ibid.). The new sinology would find in each locality a 'fragmentary view': an eclectic synthesis in villagers' minds of items from and aspects of an available fund of practices and beliefs. Freedman seeks to turn our attention from village synthesis to the common fund itself, to the
A Chinese Religion Exists
designation of its constitution and commonness and thence to the description of a common 'language' with local 'idioms'.
The distinctiveness of the 'vast field of apparently heterogeneous beliefs' and 'equally enormous terrain of varied action and association' 'lies in a basic stock upon which complex social and intellectual life works and elaborates variety' (ibid.: 21). Can that word 'stock' be free of its appropriateness to a sense of origin, to ancestry, as well as to a store of material? We are in any case dealing with the self-evidence of etbnos, which carries the promise and sets the task of finding a whole in the rules of meaning and the order of organization. But it is Freedman's rigour in facing the challenge of that task which demands attention.
His earlier work on race relations in Malaya and Singapore, furnished with the then newly developed notions of plural society, raised the problem of providing a positive designation for the different ethnic groups—one of which, of course, was Chinese. The two books on lineage organization in south-eastern China (1958 and 1966) can also be seen as designating the ethnicity of overseas Chinese in at least two ways. Lineage villages were the 'regional peculiarity' (1958: 1) of the area of China from which the overseas Chinese emigrated. Other peculiarities of the region have been named: spirit mediumship, the efflorescence of temple building, certain temple cults, certain secret societies—all of them to be found in the emigrant settlements. But Freedman's work gives pride of place to lineage organization, a decision that was reinforced perhaps by the fact that lineage is a form of organi2ation based on the same principle as that by which 'origin' itself is traced.
The belief in common descent and the principle of identification or rather of differential identification of more or less distance from common origin or place of origin—constitute the core of the definition of the 'ethnic'. The sharing of selective principles by which contemporary social distance is calculated by an ego together with what is entailed by the calculation—property and power—defines an ethnic unit. It is as well to remember that Freedman's as distinct from de Groot's ethnography contributes to an ethnology which has abandoned evolutionary origins in its objects of investigation. It has redefined origins as social centricities with their 'own' principles of organization, and this gives belief and religion a new importance. For principles are to be treated as beliefs, objects which can only be represented and so must always be the subject of the tools of inferential approximation from representations to their assumed empirical subjectivity. The tools of this social ethnography have thus been participant observation and the modelling of models.
Freedman's models are, typically, models of both institution and belief:
The models feed on the evidence as I know it; they are not intended to masquerade as substitutes for evidence. They may turn out to be bad models as we come to know more and more about China, but they are
15 o    Stephan Teucbtwang
intended as hypotheses about some significant features of Chinese ritual life and about the connections between that life and characteristics of the institutions of family and kinship. (1970a: 164)
These are models claiming to be more than theoretical constructs providing the experiments of further investigation. They are also approximations to a peculiar ethnographic evidence, an evidence (observed) of inner (unobserved but also objective) connections between ritual and institution, part of a connectedness which is the China eventually to be known.
Furthermore, they are 'outside' models of an 'inside'. Insisting upon describing Freedman's work as ethnography and his conception of China and Chineseness is to insist that it and ethnology do not escape the problems of the doctrine of subjectivity.
In 'Ethnic Puzzles' (1975 b), Freedman seeks to pin down ethnicity in this way. In this review he upbraids some of the authors of Urban Ethnicity (Cohen 1974) for lack of a clear distinction in identifying ethnic groups between an outside definition and the categories used by the subjects of study. He recommends, instead, definition according to a categorisation of insider modes of grouping. One main category is that of 'quasi-grouping' defined by contrast with structured grouping (1975 b: 60-1). Ethnic groups are, Freedman thinks, best restricted to one kind of quasi-group, namely to 'entities that conceive themselves such in virtue of their common origins' (ibid.: 57). Ethnicity would then be a recruiting principle within an order of many kinds of differentiation of groups and individuals, and of many kinds of structural aggregation into groups whose rules are less chosen. Objective as this categorization may be by its correspondence to observed distinctions 'in' all societies, nevertheless the major distinction is dependent upon a categorization of subjectivity: this is involved in his distinction between 'recognisable structure' (my emphasis), which can only mean observed from outside though not necessarily conceived by the insider, and 'active grouping', dependent upon 'categories formed by the insiders' (ibid.: 60). The difficulty is that both the outside and the inside, both the structure of involuntary aggregation as well as the categories of active grouping and differentiation, are taken to be independent of but necessarily also present in the subjectivity of human individuals. They are both 'of and 'in' a greater entity, such as that named China, which has been demarcated 'by [as he puts it] an outsider observer (the anthropologist or sociologist)' (ibid.).
Chineseness would then be a combination of structures and categories. The structures, in the jurisprudential tradition of social anthropology in which Freedman worked, would be rules or principles governing the formation and constituting the continuity of groups and institutions, while the categories and values governing the formation of equally observable, named, but less formally defined, aggregate entities are the
A Chinese Religion Exists    i; i
rules of identification and differentiation. The latter are the internal designation of subjectivity, the parsing of first, second and third person, inside and outside, and comprise the self-identification of the greater entity. But both structures and categories are participants in another 'inside', which is the way the object of anthropological enquiry is designated. All observations are thereby treated as representations of that 'inside'. Analysis of structures is still a model of the inside, and analysis of categories is a model of the insiders' models.
Evidence of their presence is first, the patterns of aggregation observed, secondly, the ritual and verbal statements recorded, and thirdly, the means of asserting and communicating rules and categories of aggregation—as codes, models and other representations. Freedman laid as much weight upon the third kind of evidence as on the other two. Hence his stress on sinology and the extensive repercussions of literacy and printing.6 Hence, too, his stress on the centralisation of the polity, its bureaucratic organization channelling and forming a unity in its own right by its rules of membership, its means of communication, and the codes it promulgated.7
These surfaces of pattern, record, ritual and writing are always to be understood as representing something else, the presence of ruling principles elsewhere to which analysis refers. All his earlier work on ritual institutions conforms to this mode of understanding, long before it is brought to the more deliberate light of conceiving a religion of China. Thus, the regulations on mourning grades in the Imperial Chinese Civil Code are read as one of a number of sets of discrepant expressions of an underlying system of ruling principles.
Of course, these codified rules do not represent for us the actual rules by which behaviour was necessarily regulated, but they reveal an official system of ideas in which close agnatic kinship was thought to set up special rights and duties standing apart from, and sometimes in opposition to, the rights and duties between man and man and between man and the state. (1958:43)
Of course they are rules, but they also express ideas which themselves are only a partial and distinctive, official reflection of the rules of which
6. See, for example, Freedman 19753:201-2: 'The printing press, vernacular literature, and the story-teller form a complex that impressed the China Coast scholars at the time of which I am writing; the vision of it was dimmed when, in the anthropological investigations conducted by Westerners, attention was concentrated on the spoken word—to the neglect of the written word upon which it in part depended.'
7. For example, in the first steps supporting his recommendation to begin with the assumption that a Chinese religion exists (1974:21), Freedman writes; 'We may start with two simple and connected propositions. First, Chinese religion entered into the unity of a vast polity. Second, it was an intrinsic part of a hierarchized society.'
i j 2    Stepban Feucbtwang
behaviour is the pattern. Such is the complexity to which representational analysis can run. And representational analysis it is:
The official pattern of mourning in this fashion portrayed the diminishing intensity of relationships between agnates according to proximity.... Although this system expressed a legal grading of responsibilities couched in the idiom of mourning, we know that people in the villages, at least, did not follow the official rules in all their elaboration. (Ibid.: 44)
The point to be made from these quotations is certainly not that the village is taken as the essence of China, nor that Freedman sets aside differences between villages and officials and the possible opposition between village and state. It is that official code and observed regularities of mourning ritual are read as (distorted) expression, or portrayal of a system, leaving uneasily the explanation of the expression (distortion) itself for some other account.
Indeed, the notion of 'expression', heavily used though it is, escapes analysis. Its variation is the reflection of the variation wrought by partially congruent principles of social order, as we shall now see.
In their partial congruence, the principles make up the whole in which any one of them, including agnation, must be understood. 'There can be no satisfactory understanding of the Chinese lineage which does not rest four-square on the study of China as a whole' (1966:159). This is not a simple whole derivable from any one key principle of organization or cultural element. More than once Freedman opposed such a conception. For example:
It occurs to me that the key term 'social stratification' may be a root cause of our failing to come to terms with some key principles of order...not as simple hierarchy but as a complex of only partially congruent rankings. (19753:208)
The principles mentioned there, principles of ranking and differentiation, were those of civility and productivity.
Earlier he had written: 'In fact there is no simple key to Chinese society' (1966:159).
Freedman's lineage study (1958) had been true to his 1966 'four-square' demand. Various principles of the organization of land tenure, differential access to administrative position, and competition, defence, and rebellion in the economies of property, sovereignty and status were considered in their effects upon the agnatic principle. Their expression in terms of the" latter is also the expression of their congruence. Filial obedience (xiao) is, for instance, in its various expositions, both an agnatic expression of an imperial bureaucratic principle and a bureaucratic expression of an agnatic, lineage principle. The expression xiao and the moral principle it represents are symbolic of the mutual support of the two systems. Its textuality or discursive condition as expression, or its ideological condition as moral
A Chinese Religion Exists    15 3
principle in any particular instance, vanish as principles of organization are read through them.
But the opacity of representation reappears in the disorder of fact. The order is on the verge of disintegration in the face of 'congeries of haphazardly assembled elements' which 'reduce the reader to a state of stunned resignation before a mass of non-aggregative facts' (1974: 20, referring to Dore's many volumes on Chinese superstitions). At the same time Freedman, reader and systematizer of disparate data 'in the dialectic of fieldwork and bookwork' (1970b: 10), confidently asserts the order they must represent, even if it is still under construction, revision, reconstruction, with no end in sight:
As out understanding of Chinese society builds up, we shall find that the more consistent a picture we make of it, the less satisfied we shall be. It will become more apparent than it is now that the variations in institutions and beliefs are of greater importance than the consistencies. (1967b: 102)
Since for Freedman consistency has not been the criterion of order, this is the opposite of an admission of despair. Freedman's picture represents the cross-cutting of systems of variation. Earlier (1958:124) the partial congruence of the systems was defined in the manner of his fellow anthropologist Max Gluckman as one of mutually supportive conflicts. Cross-cutting, they give prominence to certain nodal points. One key node in this instance was the role of lineage gentleman/official sanctioned by the principle of filial obedience (xiao), through which lineage and village alliances and feuds were regulated and a certain peace under central administration maintained with and in spite of local self-rule.
From Ritual to Religion
Freedman's work of complex systematization, discerning the play of different and often conflicting rules of organization in the data given by the field and the Chinas of past construction, explains and singles out certain of those data. The nodes of cross-cutting rules were also nodes of heightened expression, and that heightening is ritual's place in the picture.
Rites are a variety of heightened behaviour. In this essay ['Ritual Aspects of Chinese Kinship and Marriage'] I try to summarise what we may know about Chinese rites in the contexts of ancestor worship and marriage, linking the poetry of symbolism and religious belief to the prose of social institutions. (1970a: 16})
Rites and beliefs constitute the self-knowledge of the systems of institutions. What is meant by 'heightened' is not explained. Nor indeed does
Stephan ¥euchtwang
Freedman anywhere reason how ritual is to be discerned from other repeated actions and other signs. We must here understand that ritual can be distinguished by elimination, according to the ways ritual and religion are commonly defined. I have already mentioned what must be eliminated in the couples which demarcate subjective inferiority: thus rituals are repeated signs and actions which exceed technical and experimental procedures, which exceed rational means to known or stated ends, which exceed explicitly governmental regulations, and which exceed prosaic or well-defined and relatively unambivalent communication. What these excesses express, after the explicit and self-evident functions and codes have been eliminated, is Freedman's concern. And it turns out to be a similar but 'deeper' version of what is exceeded: that which lies within the conflict or non-congruence of the explicit rules and meanings.
A notable characteristic of Freedman's first published account of Chinese religion as a separate topic, thirteen years earlier,8 is the closeness of the link he finds between institutions and rites, symbolism and belief. Each set of rites and signs indicates rules of grouping and differentiation and is linked to other sets through their functioning, even instrumentally.
The system of ancestor worship practised in south-eastern China conformed to the properties of the model of ancestor worship set up by anthropologists. That is to say, the lineage as a whole and its component segments expressed themselves ritually by using ancestors as the foci for cult groups. At different times different agnatic units within the localised lineage came into religious play. But as soon as we examine the manner in which these different units were brought into existence and the way in which they operated we notice how the unequal distribution of power and status within the segmentation was given a ritual expression. (1957 Lecture notes, p. 3)
The ancestors in the halls were instruments in a struggle for status conducted in accordance with certain rules. (Ibid.: 6)
'The Chinese belief in the three-fold division of the human soul' (ibid.) between purgatory, grave and ancestral tablet thus expresses three distinct kinds of organization. Purgatory is an expression of central bureaucratic organization; the grave and its geomantic siting are expressions of competition for status, power and the establishment of lines of descent; the tablets and ancestor worship are expressions of established households and lineage segments.
In later work, the links are elaborated and redefined as expressions of distinct kinds and limits of jural authority and of struggle for status and wealth:
1. Households and economic co-operation among them are focused on recent ancestors, tended by women and recognized as the
8. This was a lecture to the Royal Anthropological Institute entitled 'Religion and Society in South-Eastern China', summarized in Man (Freedman 1957).
A Chinese Religion Exists    i; 5
subjects of spirit-mediumship and geomantic competition. Larger organizations and alliances of lineages and their local segments are focused on more distant ancestors, distinct in halls, ceremonies excluding women, and by the absence of divination.
2. Territorial administration (less of a concern for Freedman) is organized, sanctioned and expressed in households with their stove gods, individuals with their purgatorial or fated souls, and a bureaucracy with its purgatorial and cosmic equivalent.
Extensions and additions on these lines could be and were made, by Freedman and others. But Freedman turned to descriptions of cosmolog-ical beliefs and associated rites which were not as immediately linkable with units of social organization as were ancestors or territorial guardians. Indeed, as Freedman noted, geomancy usurped the 'pride of place' in 'ritual matters' which ancestor worship had held in his previous work (1966: 118) because of its cosmological content.
Cosmology promised to be more than one of the heightened expressions or foci of different kinds of organization or principles of jural authority and property. It promised more even than the expression of points of their combination, for it held out the prospect of revealing the principles underlying the combination of those expressions, thus disclosing the 'totality' itself, the very basis of the rules of Chineseness.
Note that to Freedman, the culminating chapter of Granet's The Religion of the Chinese People, 'the climax of the book' as he called it (1974:34), is the chapter on the official religion. There Granet brings together the peasant and noble, country and town life of 'feudal' society, the 'agnatic and primogenitary kinship system and the concomitant elaborations of the ancestor cult, the worship of Heaven, a dynastic and official cult superimposed on the agrarian and ancestral cults' (ibid.: 35). Together they are 'worked up into beliefs and cults serving the needs of an imperial state and its functionaries, the literati.... We have arrived at a synthesis in which beneath the surface variety and literate sophistication there lie a few simple basic religious ideas, a heritage from the peasant past' (ibid.: 34).
Whether or not they constitute the past attributed to them is a matter for historians. The principles are, with a few modifications and some additions, those which Freedman had already inferred from his own socio-anthropology of China (e.g. 1966:152). Yin-yang f^^, according to Freedman, is the fundamental concept of Chinese religion for Granet. It is also fundamental for Freedman, though in his new sinology it figures as an organizing principle of ritual and belief. It unifies, or at least articulates, agnatic beliefs with calendrical, agrarian, territorial, and terrestrial beliefs and rituals. Yin-yang organizes ancestral ritual into two aspects, that of graves both immediate and distant and that of tablets both in home and
15 6    Stephan Feuchtwang
hall (1970a: 168-9). On another scale of organization, it articulates the relativity of ancestral origin itself by differentiating the spirits of the dead into shen if, which ateyang, and guife, which arejrá; my forebears, who are shen (ancestors), and the forebears of others, who are gut (ghosts):
Yin-yang makes a primary distinction between ghosts and ancestors; it goes on to discriminate between two aspects of one ancestor: his status in the tomb and his status in the shrine. (1967b: 86)
Most important, yin-yang is a concept fundamental to cosmology, a Chinese universalizing and totalizing belief. That alone, and nothing else, makes it religious. It vouchsafes the promise of discovering in China a positive principle of Chineseness.
Conclusion
Between Freedman and Wolf, the only point at issue is whether they are dealing with one big Chinese ethnicity or a number of related but different ethnicities. The religion whose existence Freedman recommends us to assume is the representation of that unity without a key principle which has always been the object of his work: the greater ethnic subject. But in the rigour of its pursuit of Chineseness, Freedman's work has indicated quite other possibilities at the margins, where the whole enterprise is at stake.
If we were to generalize Freedman's treatment of Chinese religion into a definition of religion as such, what would it be? Certainly it would have to be treated, with Freedman's own caution, as a term of art, a definition that has directed his work or, rather, a way of summarizing my exposition of what has directed his work under that heading. The definition is, then, that religion is a heightened expression of the ruling principles of a social order. That social order, as order rather than as the observable material of statement, actions and associations, consists of ideas and forms. Freedman did not psychologize. But ideas and forms were, as I have argued, understood to be an inner reality which is mental. I have argued that Freedman's modelling of 'insider' models, if not conceived according to psychological theories of cognition or of drives to find order, does negotiate a space of mental processes, of 'knowing and apprehension', 'experience and belief (1979). There is room here for theories of human psychic processes, of cognition, perception, image formation, fantasy and projection, to do their work of ordering and disordering. But I have followed Freedman in leaving them aside, including with them that structuralism of Levi-Strauss and other universal theories of mental
A Chinese Religion Exists
functioning, and attending to a more historical ordering which is at once social and mental. It is social precisely in not being true of a universal or an individual human being but of an 'ethnic' peculiarity. In Freedman's work, religion belongs to this social mentality: the expression and communication of ideas and forms of social order.
Now suppose we were to take Freedman's own discipline of dissatisfaction with a picture of consistency and take his stress on the importance of variations in institutions and beliefs further, no longer modelling from their evidence a basic social order represented or expressed in rituals and symbols. Some order there is. But we need not expect a basic order that orders the variations of everything Chinese and is expressed in symbols and principles. It is only demanded by a philosophy of the social and the human, which can be abandoned without thereby abandoning hope or forsaking reference to any order of facts. Without begging any questions of what they mean, we still deal with the facts selected by the term 'religion'. Magic, sacrifice, worship, witchcraft, ritual and myth are, as Freedman wrote (ibid.), categories of fact in the anthropology of religion. But Freedman's work on Chinese religion confronts the possibility of having to admit that ancestor worship, divination, rituals and representations of gods—to select only those 'religious' subjects to which his work on China was devoted—could not easily and perhaps not at all be bound into a.5ingle totality of 'Chinese' religion. And if this admission is faced in the case of China, must it not also be faced as a possibility in the case of any ethnography? And if that totality embraced by 'religion' in each of its social contexts cannot or should not be assumed, then what it links should not be assumed to have any coherence. Coherence, systematicity, and the disposition of similarities and differences defining a difference between what can be described as 'Chinese' and something else, all these must remain open to being found, or not found', and not finding them is not a sign of anthropological failure.
Suppose, then, that we forsake the embracing security of the category 'religion' and take up Freedman's analogy with language and idiom. We are still left with whatever has been selected by the term 'religion'. Magic, sacrifice, worship, witchcraft, ritual and myth are all representational facts in that they signify and are discursive. Their evidence, if I may so put it, is also evidence for the empirical study of representation. Suppose, then, in taking Freedman further we no longer model an inner social mentality but instead adopt a new rigour of describing the surfaces and means of representation as such. We would refuse to assign a prior reality to ideas and forms. We would take ideas and beliefs to be in the processes and mechanisms of representational formation, not using or being obscured and distorted or just reflected by them.
Instead of the assumption that a Chinese religion exists, and instead of the question to which it responds (does a Chinese religion exist?), a new
15 8    Stephen Feuchtwang
question now presents itself. What are the properties differentiating the 'variety' of representational surfaces themselves? The question is prompted not only by Freedman's analogies with 'idiom' and 'translation', but also by the work which has been done on signification, symbolizing, ritual, ideology and discourse into which the evidence of whatever is described as 'religion' falls.
Much of this work is linguistic, and that raises immediate problems of analogy and similarity, since the units of meaning described as religious are in orders larger and differently constructed than those of linguistic signs and statements. Ritual, myth, liturgy, text, rhetoric, oratory, each suggests a descriptive vocabulary appropriate to and already applied to representations selected by the term 'religious'. None has prior claim to. govern the others.
The ethnography of 'China' which claims entrance into the general literature of social and cultural anthropology was begun long ago, but it will have to go further in adapting and transforming such ways of studying the representational forms themselves of 'Chinese religion'. As long as the study of Chinese religion is still bound into a reading which wishes to see through them a China or many Chinas as yet unknown, the ethnography of China will remain parochial.
We have still to ask what, if any, kind of representational form is selected by the term 'Chinese religion' in Freedman's work. There are two answers to this, one concerned with forms of representation, the other more concerned with their character.
Freedman selected the form 'ritual' and described it as 'heightened expression'. I have already interpreted this to mean what ritual often means as an adjective in anthropology and elsewhere: repetitive and excessive. That which is repeated, and exceeded by being repeated, is other forms of representation and communication—forms of address, of deference, of gift, of transaction and of feasting which also accompany and are marked and timed into occasions by the ritual forms. The meaning of the excess, the meaning of this quality which is called 'religious', could simply be that: the marking out of spatial and temporal forms. Firecrackers, for instance, are markers, noisemakers that punctuate approaches and separations (Freedman 1967a). Beyond is only what is marked by ritual, namely that which is exceeded and referred to by repetition and ornamentation.
'Heightened' can also be understood as the marking of a point of multiple reference across discourses; focus, certainly, on a particular form of communication or of social organization, but also a conjunction with others underplayed or overplayed.
Take as an example the symbol and term which Freedman lighted upon as a principle of Chinese totality itself. In the project suggested now, jin-yang is only whatever it is found to elaborate or organise. That it can refer to a great deal more does not, moreover, mean that it refers to all of
A Chinese Religion Exists
it in each iteration, nor that each iteration partakes of the same essential meaning. What is different in each instance of its use and reference is to be assessed in each case. Yet it must be true that each timcyin-jang is used, it has its own effect, a dualism; an immediate contrast and complementarity of influences running in cycles is introduced when either one of its halves is stated and thereby a framework forming terms and references connected to it. But the difference between each usage has as great a claim to priority as the repetition of the same terms, unless we are to be ruled by some etymological rectification of terms. Nor does its use predict how long that dual form lasts in its effect upon further events in a discourse or a ritual chain, how much it structures and what more than a complementarity it suggests while the effect of its use lasts and before another ordering principle (such as a principle of precedence) is introduced.
What is repeated, what contrasted, what ambivalent cross-allusions are made can be specified. The excess is thus in their specific conjuncture, the overtones it produces in their effects upon each other at that point.
Further, and beside references away from ritual, there is also the tracing of references to distinguish what might be called 'an economy of ritual information'. The phrase was in fact suggested to me by Freedman himself for my own work on Chinese religion (Feuchtwang 1975). By this is meant the investigation of what acts or objects or phrases or liturgical themes and ceremonials do any observed rituals refer to others, and by means of what accessible specializations and roles. This will involve tracing any one ritual to textual traditions, or to historical myths, or to other ritual occasions and other disciplines of precaution and special activity (of fasting, of feasting, of retreat, of magic). Thus, in short, a pattern of ritual discourse and a lexicon of forms and their references would be described in all its particularity.
What, next, is the character of those representations Freedman selected as 'religious'? Cross-cutting principles of order, ruling principles and forms of association, were marked out as 'key' and 'focus' in Chinese religion according to Freedman. It is difficult to decide whether this was a result of an English anthropological tradition or some other predilection, or whether it was more the effect of the Chinese materials themselves that made Freedman characterize them as jural. In any case, the jurisprudential analogy could fruitfully be taken more literally. In the rituals of 'Chinese religion', what are repeated can be interpreted as commands, claims and appeals to adjudication or just to outcomes and to engagements in chains of reciprocity or of hierarchical contract. The rhetoric and mythology, the gesture and costume of Chinese religion are those of historical mythology and more particularly of imperial rule. Their references, that which they exceed and repeat, are the representational forms of greeting, address, transaction and plea which actually also take place in commercial and political relations but sealed, started, or replenished on ritually marked
Stephan Feuchtwang
occasions (see Ahem 1982). Furthermore, I have found that close attention to the ritual forms themselves—not to a politics or a commerce which is read as their basic material, but to the material of the rituals themselves—reveals (for instance, in the organization of festivals, in the cults and their paraphernalia of ornament and performance) a number of forms and principles of organization (Feuchtwang 1974, 1977)- But Freedman's jurisprudential analogy suggests going further than this and using concepts from the study of law to interpret the statements and forms of Chinese 'religion'. The rituals of pleading and commanding would then be read as forms of recognizing and defining capacities of action and effect, wrong and benefit. What social forces of action and through what positions and limitations those capacities could be claimed is for further analysis. But to treat 'religion' as a discourse of recognition is, I think, powerfully suggestive.
The materials provisionally collected under the heading 'Chinese religion', and many of the analyses of rules made by Freedman, would be a field of jural signification defining subjects, not the law of a social subject. No doubt new concepts would be required of the jural and of religion to designate this Chinese field, a reconceived 'law' redefining 'religion'. These are some of the challenges thrown off by Freedman's work. The assumption of unity with which he worked latterly was, I have argued, a necessary assumption given in the notion of religion. Its particular ethnological usage could alternate between Freedman's conception of total (complex) unity of Chineseness and Wolf's idea that there are particular unities of local Chinese ethnicities. Were the frontier challenges of Freedman's work taken up, the notion of 'religion' would break up and both Freedman's and Wolf's assumptions would give way to discontinuities, conjunctions and effective combinations of juristic rules and agents in an economy of ritual and discursive forms of statement. Working that out is an intellectual heritage, and no doubt one to be disputed. But it is blocked by the notion of 'ethnicity' and by its carrier, the notion of 'religion'.
References
Ahern, E. 1982. Chinese Ritual and Politics, Cambridge: Cambridge University Press.
Cohen, A. (ed.) 1974. Urban Ethnicity, London: Tavistock [ASA Monographs, no. 12].
A Chinese Religion Exists
Dumont, L. 1966. Homo Hierarchies, Paris: Gallimard.
Feuchtwang, S. 1974. 'City Temples in Taipei Under Three Regimes', in M. Elvin and G. William Skinner (eds.), The Chinese City Between Two Worlds, Stanford: Stanford University Press.
... 1975. 'Religion and Religious Change in a Market Town on the Mountainous Rim of the Taipei Basin, Northern Taiwan', University of London: Ph.D. thesis.
1977. 'School-Temple and City God', in G. William Skinner (ed.), The City in Late Imperial China, Stanford: Stanford University Press. Freedman, M. 1957. 'Religion and Society in South-Eastern China', Man, Vol. LVII,pp. 56-7.
... 1958. Lineage Organisation in Southeastern China, London: Athlone Press [LSE Monographs on Social Anthropology, no. 18].
1966. Chinese Lineage and Society: Fakten and Kwangtung, London: Athlone Press [LSE Monographs on Social Anthropology, no. 35]. 1967a. Rites and Duties, or Chinese Marriage [Inaugural lecture, LSE], London: G. Bell.
1967b. Ancestor Worship: Two Facets of the Chinese Case', in M. Freedman (ed.), Social Organisation: Essays Presented to Raymond Firth, London: Frank Cass.
1970a. 'Ritual Aspects of Chinese Kinship and Marriage', in M. Freedman (ed.), Family and Kinship in Chinese Society, Stanford: Stanford University Press.
1970b. 'Why China?', Proceedings of the Royal Anthropological Institute for 196p, pp. j-ij.
1974. 'On the Sociological Study of Chinese Religion', in Arthur P. Wolf (ed.), Religion and Ritual in Chinese Society, Stanford: Stanford University Press.
1975a. 'Sinology and the Social Sciences: Some Reflections on the Social
Anthropology of China', Ethnos, Vol. XL, pp. 194-211.
1975b. 'Ethnic Puzzles', Jewish Journal of Sociology, Vol. XVII, no. 1, pp.
; 5—66 [Review of Cohen 1974].
1975c. 'Introductory Essay', in Granet 1975.
1979. 'Social and Cultural Anthropology', in Main Trends of Research in the
Social and Human Sciences, Part 2: Anthropological and Historical Sciences, The
Hague: Mouton, for UNESCO. Granet, M. 1975. The Religion of the Chinese People (transl. and introduced by
M. Freedman), Oxford: Basil Blackwell. Renan, E. 1947 [1863]. Vie de Jesus, Paris: Calmann-Levy. Spinoza, B. 1951 [1670]. A Theological-Political Treatise, New York: Dover. Wolf, Arthur P. (ed.) 1974. Religion and Ritual in Chinese Society, Stanford:
Stanford University Press. Yang, C.K. 196 i. Religion in Chinese Society, Berkeley: California University
Press.