In Religion as a Cultural System Clifford Geertz suggests a definition of religion as "A system of symbols which acts to establish powerful, pervasive, and long-lasting moods and motivations in men by formulating conceptions of a general order of existence and clothing these conceptions with such an aura of factuality that the moods and motivations seem uniquely realistic." The bulk of the essay is devoted to explaining this definition and illustrating the explanation with examples from various religious systems. This is an extremely thought provoking essay, and it gives one a good idea of why Geertz has such stature among anthropologists.
Turner's essay Colour Classification in Ndembu Ritual is a study of the role of the red-white-black colour trio in Ndembu ritual life (and more generally in ritual systems in Africa and around the world). He restricts himself to a descriptive, down to earth approach, except that towards the end of the essay he proceeds "to throw caution to the winds" and goes completely overboard in theoretical speculation. He starts with a claim that dyadic and triadic relationships are fundamental in human relationships, which is something that (while plausible) is hardly incontestable. He then concludes with "...the three colours white-red-black for the simpler societies are ... abridgements or condensations of whole realms of psychobiological experience involving the reason and all the senses and concerned with primary group relationships. It is only by subsequent abstraction from these configurations that the other modes of social classification employed by mankind arose." In short he claims that colour symbolism is fundamental to human social classification!
In Religion: Problems of Definition and Explanation Melford Spiro presents a definition of religion as "an institution consisting of culturally patterned interaction with culturally postulated superhuman beings". He presents a reductionist and individualist view of religion, where causality is fairly strictly individual => religion => society. He argues (against functionalist explanations of religion) that an unintended consequence of religion cannot be a reason for its existence. Now while this is clearly true of religion considered as a conscious behaviour system, it hardly seems true in general. One alternative possibility is that something akin to biological natural selection applies. Random (in the sense of sensitively contingent on history) variation in religious systems (and other social structures) seems inevitable, and if societies without religious systems of certain types disintegrate due to lack of social stability then in some sense it makes sense to say that religions exist because they produce social stability. Such a theory might also be able to explain the variety of religious systems in existence, something Spiro stresses needs explanation.
The last two essays are slightly shorter. R.E. Bradbury's Fathers, Elders and Ghosts in Edo Religion is a structuralist study of the role ancestor worship plays in Edo social structures. The basic idea is that relationships between the living and the dead and between different classes of ghosts and spirits are projections of the social organisation of the living. It is suggested the oppositions considered in the essay are just part of a more abstract order which can be used to classify a much wider range of human experience.
Edward H. Winter's Territorial Groupings and Religion among the Iraqw is only peripherally about religion. He studies an ethnic group whose social organisation is based on community and territory rather than on the lineage system as in most African ethnic groups. The structural principle of Iraqw society is spatial contiguity rather than kinship, so the members of a lineage are scattered throughout the population and can move around freely. The organisational units above the village are also geographically organised, and it is religion that provides their coherence, with councils responsible for organising rituals for rain control. The situation among the Iraqw is contrasted with that among the Amba, who have a more "traditional" lineage system. It would be interesting to see if the Iraqw have coped with modernisation better than other ethnic groups as a result of their organisational differences.
I have tried to give some idea of the content of the essays: suffice it to say that they are all extremely thought-provoking, whether one agrees with them or not, and that it is easy to understand why this collection has been reprinted so often.