The ancient mind Elements of cognitive archaeology Edited by COLIN RENFREW and EZRA B. W. ZUBROW 194 3 CAMBRIDGE 9 UNIVERSITY PRESS6 The archaeology of religion COLIN RENFREW ‘Any attempt to encompass the archaeology of mind must inevitably consider the archaeological approach towards religion. For ifthe archacology of mind, as envisaged in chapter I, may be considered in terms ofa Series of functions Of the symbol, of various ways in which symbols may operate, the ole of symbols in coping withthe unknown and withthe supernatural is surely one of the more significant (see Renfrew and Bahn 199%: 358-63). But there isthe danger here that we may cary to the inguiry our own ‘culturlly-eneapsulted, and therefore perhaps stereotyped, View of what religion is. Though our acquaintance, in the Fist instance, withthe grat religion ofthe Book (dais CCuistanty, islam), allof which proclaim a unitary deity we undoubtedly begin fom a very special viewpoint. Even some acquaintance with otber great, contemporary faiths such asthe Budchst, Hindu, Jain and Zoroastrian, serves in some ways to reinforce the impression of coberenlycoified (od thus literate), authorative systems of belie, operating often in an urban context A preliminary knowledge ofthe religious systems and the pantheons of Ancient Egypt Greece and Rome might, a first, reinforce this view of tet. ‘sed, wel-delineated and formalized structures of bei. Clearly, however, the studies of cultural anthropologists have much o tellus aboot the religious systems of non-urban societies and of social groups operating on a basis of band ot taba organisation, so that some ofthese preconceptions can be counteracted and the eflecs of literacy discounted ‘A more serious difficulty pchaps accompanies our very concepaualization of ‘religion’ itself, asa distinguishable, and in some senses separable, Feld of human activity. For we shal son noe that, from the standpoint ofthe archaeologist, Feligious activities aze potentially open wo observation only ‘when they might be idetifabe as religions by an observer AC the time in question, Places set aside for religious observances and objects used specifically for cult purposes ray, in favourable circumstances, be recognized a such. Such identification is mich less easy when the locus of religious sevity has a whole range of ater functions, or when the artefacts used thee algo have other, secular uses The problem ofthe ‘embeddednese™ of cule activity within the other activites of daily life is thus very real ene. And {vst as economic anthropologists warn us that the economy in simpler, non-state economies is often inextricably ‘embedded within the matrix of the social organization, £0 might we anticipate that the same could apply to cult ‘observance. The very term ‘religion’, conceived a= a ‘separate dimension or sub-system of the society, could thus prove tbe something of « misconception, even among doce ‘communities where the supernatural plays significant role in shaping the thoughts and ations oft individuals, These, however, are inescapable constants atleast we ty fo be aware of the, We should therefore concede that in many Societies the religious life is more varied and more widespread than we ight realize, especially if thas this quality of embeddedness. Inall temps to investigate the erly past there isthe risk that we frst conceptualize, seting up a whole series of categories of our wa constuction, and then order our data (our observations bearing upon the past) in terms of such categories. The past then presented in these terme, and itis easy to assume that our description i telling us about the ‘way the past was andthe way it was ordered. In some cases, however, all that we are Sceing is a reflection and an exemplifiation of our own a priori categories. Such criticisms have, for instance, been made of the tem “chiefdom, much used for a while among evolutionary anthropologists, and sill useful among archaeologists, Butt has been pointed ou, with some justice, that it can be a ‘cumbersome exercise to set up elaborate criteria by which a chiefdom may be recognized, and then to spend much time and effort arguing whether this culture rtat society ito be regarded as chiefdom agains the standard of those ctr. ‘The wltimate moral must be that such classifications are not useful in themselves. Classifications ar of value only if they fae put 10 some use once they are established. ‘These ‘cautionary thoughts are perhaps easier to formulate than they ae to apply in practice, but right a the outset of our con- sideration of the archaeology of religion, it should be scknowledged thatthe very ute ofthe category ‘religion’ ‘evitably influences same aspects ofthe discussion. Religion isnot an easy term to define. But it cleanly implies Some framework of beliefs. These cannot, however, be48 ColinRenfew resticted to general philosophic belies about the word or out the way it works. They must late to forces which are fot merely those ofthe everyday material world, but which so beyond it and wanscend it ‘The Shorter Oxford Dictionary ofters one convenient efition (Onions 1973: 1978) for religion: “Action or ‘conduct indicating a belief in, or reverence for, and desire to plese, divine ring power... Recognition onthe part of ‘nn of some higher unseen power as having contol of his Gostiny and as being emted to obedience, reverence and warship” This convenient definition has many merits, bat rot al ts components may be of universal validity. For insance, there ae some oriental belie systems (such as that following Confucius) which are generally recognized as religious but which avoid specific divinities and where such powers as are postulsted ae immanent, not readily to be Separated from other aspects ofthe word with which they ae tobe astocated, “To speak ofthe “supemarural” in such a case might be misleading i it were taken to imply a belie! in spirits or ‘ther separable entities, Bu the Shorter Oxford Dictionary ‘einition for ‘supernatural’ is perhaps a broadly accepable ‘one (Onions 1973: 2193) “That is above nature; wanscending the power ofthe ordinary course of nature’ Transcendence doesnot necessarily imply separation These are important and basic points. Durkheim, for instance (1965; 47) was able to define religion without reference tothe supernatural although he could not avoid the term ‘sacred, Geet (1966: 4) fers definition which indeed avoids the superatural andthe sacred, but which is solackng in focus that could apply to secular etal oreven tothe system of values which i wed t uphold a monetary ‘cermomy. Such & definition lacks any sense of what most surely be a component of any religion: the individual religious experience Reigious experience ‘Cental to the notion of religion is that of personal experience forte individual which Seems to him or her not only imporan, but ofa larger significance. This sa feature teeflly discussed by Rudolf Oto (1917) in a work which sll makes valid points, where he laid emphasis onthe sense ofthe numinous which he viewed as central 10 religious fewperience. It cannot be escaped that when we allow ‘outelves to speak of the rligons of early societies we are ‘iking something ofa cross-cultural assumption, namely that there was indeed Some variety of numinous experience cexjoyed by the members of those societies, some sense of mystery and of external, non-human power. We donot need toasser that such a power has any real existence ~ that i a mater forthe indivdval's religious beliefs today. Butt talk of the archaeology of religion presupposes that religious ‘experience was availabe then as now. Ths is a pot wich, it would be very dificult to demonstrate, although sich an ‘experience is cenainly perfectly plausible as a motivating force for some of the symbolic monuments which we may observe from the past But the existence of such an expe fence in the past seems tobe an astumption Which the sade of early religion has co make: there is 2 uniforitaian ssumption there which needs tobe recognized “This is 2 question which [believe was largely avoided by Durkheim, and then less subly (and less succesfully by Geer inthe dgcussons cited above. Both avoided placing such religious experience as a central feature of ther 50 Colin Renfrew to se tat there is mh inthis idea. Whether man is made {nthe image of God or vice vers it is certainly the case that the deities of many human societies have been conceived of 2 anthropomorphic, although animal gods (theriomorphic) fre at lest as-common. The notion of a hierarchy of divinities is one which comes more easily toa society which fs iteelf hierarchically structured, for instance to a state society, It is ificlt to imagine such a feature in an cgaltarian society organized at band level ‘The functions and consequences of religion ‘Various thinkers have considered religious systems nt from the standpoint of the content of thet belief structures nor oftheir communal behaviou, but in terms oftheir conse “quences for society as a whole, Within the Marxist view, for instance, with ts empbasis upon class antagonism, religion is viewed as a means, developed by the élite, for the ‘manipulation of the masses. Within most state societies itis ined the eae that the ruler (thé “king’) has 2 special place in relation to the leading religious specialist (the ‘chief priest’) and in many state societies the two offices converge {the ‘pries-king), sometimes with the deity ruling through this head of state (4 theocracy), oF a least inspiring and sanctioning his ations (he “vine right” of kings). Marist, fchacologsts have also applied such ideas 10 societies fenerally regarded as essentially egalitarian, where, for instance. the elders ate seen as using tligios beliefs to ‘ensure their own favoured and privileged position in celaton to younger members of society (Tilley 1988). Avery different view of religion is taken by anthropologists ‘working broadly inthe “fonctionalis’ wadtion of Malinowski, for whom religious beliefs and rituals are seul in governing and regulating varios aspects of the Social system (eg. Reppaport 1971s). The earlier ‘function alist view saw religion as useful in ensuring the smooth functioning of society by ensuring some considerable degree fof community of belief, some acceptance of the social System and hence some general social solidarity among members of the community, More recent work, notably by Rappaport. would accord religious beliefs a further, more active role, in bringing into play mechanisms of a broadly homeostatic nature whose timing is governed by situa "A further, broadly evolutionary, perspective has been proposed by Lachmann (1983) who contrast the variability inall human behaviout with that among species where quite complicated social behaviour is genetically determined, for instance among bees, In his perspective there are enormous alivantages in cultural evolution, precisely inthe special human ability 10 lear, and thus to change the whole behaviour of society in the space of just a couple of _Benerations, But there are penalties too, in terms of lack of stability of behaviour and in the isk that valuable infor- tation, no longer stored in the genome, may become lost between generations, I ie here that the devices used by religion for the very long-term storage of information may be particularly useful, For, as Lachmann stresses, itis thvough religion that “ancient wisdom’ is most effectively sored and transmitted. Mach ofthe repetition assocsted ‘vith elgious ritual may be regarded as “redundant” inthe information they sense, and secure transmission i thereby more adequately assured. It is the case that the oral ‘tansmission of religious knowledge has been shown t be ‘enormously effective. The Hyms ofthe Rigveda recorded in archaic Vedie Sanskrit, were preserved orally for many Centuries before being et down in writing about the fifth century BC ata time when classical Sanskrit was already in tse. Mach of their vocabulary was by that time not well tnderstood, bt ater scholarly study as shown the acuracy of transmission to be remarkably high. This gives supporto Lachmann's view that a coherent religious system, wih it high survival value, confers selective advantage upon the society which possesses it general, the wider effects of religious beliefs have been litle considered by archaeologists. Yet clearly frameworks ‘of religious belief provided strong motivational contexts for many aspect of the behaviour of ancient societies. This has teen well documented, for instance, for the case of the ancient Maya (Schele and Mille 1986) and the role of sacred onceps, bth inthe development of urbanism and in its Specific forms, has been emphasized by anumber of authors, including Wheatley (1971), For the traditional Marist historian these may be mere epiphenomena, where the superstructure follows the economic infrastructure, but neo ‘Marnst thinkers are inclined to ive greater causal weight 0 ideational factors. For the functionalist there is a similar inclination o see symbolic factors as contributing to greater cetfcieney in the culture system and thus having an adaptive ‘Value, but perhaps not much more than this. Such a view Seems related tothe Binfordan notion that ieatonal and feligious aspects are akin o“palacopsychology” (see chapter ‘sand that more basic causal factors reside inthe ecological relationship between humans and ter environment “Today, however, within the framework of cognitive: processus archaeology, such assessments seem deficient. I {5 no longer sufficient to see the ideational component of, carly societies (including their religions) as simply some Supersirctaral reflection of the more substantia) ints ‘Structure albeit one that is eaily subverted by the dominant lite in their ceaseless application of the class war. Not an religion be seen a6 purely and simply a device for promoting efficiency, or even for lengtening the memory Span of society within an evolutionary context. The causesvt valuable infor: ray become lst devices used by ‘information may mast effecvely tition associated redundant” in the missions thereby ‘se that the oral been shown to be Rigveda, recorded ¢toaly for many ng about the ft ‘rit was already in tat time not well, own the accuracy fe gives suppor as system, with its ‘vantage upon the s beliefs have been leary frameworks ional contexts for societies, Tis has ve the case of the ithe role of sacred ‘anism and in its numberof authors, raditonal Marxist fomena, where the strtue, but ne0- tercausal weight to there 8 a similar iting to greater Tain an adapave this. Such view ‘that ideaional and tology” (se chapter ide inthe ecological ork of cognitive ts seem deficient. I ‘onal component of 1s) simply some 2 substan infra ‘ed by the dominant the class war. Nor imply 2 device for hening the memory context, The causes The archaeology ofreligion SI of religious change are not well understood sed, indeed, have scarely been adresse i any coherent way. Even the notion of the evolution (in the sense of gradual. fendogenously-produced change) of religious belils and practices his searcely been developed. Iis often, in practice, ‘widely assured that any new religious fom has an ouside ‘origin; the oc factors favouring or resisting change have ‘ot yet beer analysed in detail, But the religious system ofan ‘aly society can no longer be considered a Secondary factor in the explanation of culture change. Recognizing religion Some of the foregoing discussion may seem a shade premature, wen tis quite evident thatthe identification and tloeidation of early cult practices fom the archaeological record isa challenging task. and th analysis ofthe belief systems which sistain them an even more difficult problem. ‘The approgsiate methodology for these lsks is only now being deveoped, and the focus of atention has so far been upon the cognition of cult (Renfrew 1985: chapter 1 Renftew snd Bahn 1991) rather than upon any attempted inference wards the bli system underying it So fr. a we shall te in the next section, this has been examined largely in te context ofthe analysis ofthe iconography of