^{page 85 note 1} Introduction to Essays in the Ritual of Social Relations, ed. Gluckman, M., Manchester University Press, 1962Google Scholar. (If I deal with Gluckman's essay in isolation from others in the book it is because one of them, that of Fortes, represents an instance of Gluckman's general point of view, while the other two seem to have no very close connexion with it.)

^{page 85 note 2} Rhodes-Livingstone Papers, No. 33, 1962.

^{page 86 note 1} I have touched on this point in a rudimentary way myself, first in The Gods as Guests, Nigeria Magazine Special Publication, Lagos, 1960, p. 70Google Scholar; and again in ‘The Kalabari Ekine Society’, Afri, April 1963.

^{page 87 note 1} In the questions I ask here, I owe much to the influence of Professor Daryll Forde, who has more than once pointed out how crucial they are in the comparative study of religion.

^{page 87 note 2} For an exposition of this important difference between nineteenth- and twentieth-century agnosticism, I am indebted to Professor Evans-Pritchard's ‘Religion and the Anthropologists ’ (reprinted in Essays in Social Anthropology, London, 1962).Google Scholar

^{page 88 note 1} Manchester University Press, 1957.

^{page 92 note 1} Chihamba, p. 92. Though the author mentions no contemporary names, it would not perhaps be entirely fanciful to see in Melville's whaling ship Pequod the Manchester School of Social Anthropology; in Captain Ahab the powerful, charismatic figure of Professor Gluckman; and in harpooneer Queequeg Dr. Turner himself.

^{page 93 note 1} The importance of translation into one's own language as a crucial phase of anthropological analysis may seem self-evident. But the implications of its importance have often been neglected. In making the search for appropriate translation instruments a pivot of my analysis, I am following the guidance of such people as Professor Evans-Pritchard, Dr. Godfrey Lienhardt, and Professor Ernest Gellner. The latter's recent paper Concepts and Society (Trans. 5th World Congress of Sociology; Washington, September 1962) is particularly illuminating on this subject.

^{page 94 note 1} Here I follow the Oxford Linguistic Philosophers in assuming that one cannot discuss problems and preoccupations as distinct from the terminology used for expressing them: change the terminology and you have changed the problems. Where I do not follow this school is in their tendency to assume that the Thomist sense of ‘exists ’ is an abuse of language and hence meaningless in an absolute sense: Thomist thought would have been dead long ago had things been as easy as that. I am implying, however, that since ‘exists ’ means quite other things in other conceptual frameworks, the problems it raises for Thomists cannot be said to force themselves upon all men.

^{page 97 note 1} Readers of my earlier article on ‘The Kalabari World-View ’ (Africa, July 1962), will see that I am here returning to the line of interpretation that I used in the earlier piece in connexion with a particular African religion. In the present article, however, I shall be concerned with the wider implications of the approach, and will try as far as possible to avoid repetition.

^{page 97 note 2} One reason why ‘Rules of Scientific Method ’ have been developed so rarely in human history may be that they inevitably enjoin a continuous readiness to scrap established models as soon as these fall below their standards. Such a readiness involves attitudes which run counter to some of man's most deep-seated psychological tendencies; and it can only be developed in certain particular social and environmental settings which offset these tendencies.

^{page 98 note 1} The Philosophy of Science, London, 1953.Google Scholar

^{page 98 note 2} Foresight and Understanding, London, 1961, chap. 5, pp. 94–98.Google Scholar

^{page 99 note 1} Though manifest regularity is almost certainly a crucial prerequisite for selection as the basis of a model, it would obviously be an over-simplification to treat it as the only factor involved. Most African thought-systems, for instance, are aware of well-marked regularities in the inanimate environment which they have nevertheless failed to exploit in their model-making operations. Here I am thinking especially of the regular behaviour of sun and moon which, of course, are noted in all African societies. The reason in this case may be that the regularities involve isolated celestial bodies rather than a whole system of such bodies, and hence do not provide structure rich enough to serve as the basis of an explanatory model. Contrast the African situation with that which prevailed in the Ancient Near East, where the regular movements of a large number of celestial bodies were noted, and where the resulting system became the basis of the explanatory model which we know as Astrology.

^{page 99 note 2} In his Greek Science (London, 1961)Google Scholar, Benjamin Farrington has given a brilliant account of the emergence of impersonal explanatory models in early Ionian thought. He shows how closely the Ionian development of such models was linked, not only to the great technological development of the civilization, but also to the fact that the ruling and leisured classes of society took great pride in these developments and were thoroughly conversant with them. They were therefore familiar with a wide range of order and regularity in the inanimate world, and it was this range that they drew upon in the creation of their models. Farrington suggests that the decline of Ionian thought is linked with the growth of a slave-owning economy, with the subsequent decline in the importance attached to labour-saving technology, and with the relegation of existing technical know-how to the lower orders of society. As this process gathered way, the ruling and leisured classes became less and less conversant with the regularities of the inanimate world as revealed by technology; and personal models began once more to dominate the intellectual scene.

^{page 100 note 1} This is an experience reported in connexion with religious systems in other parts of the world. Thus in his Political Systems of Highland Burma (London, 1954, p. 14)Google Scholar, Edmund Leach says of a Kachin who claims to be killing a pig and giving it to the nat spirits: ‘It is nonsense to ask such questions as “Do nats have legs ? Do they eat flesh ? Do they live in the sky?”’ From the fact that such questions are nonsense to the Kachins, Leach concludes that they do not take statements about the nats as literally true. But since it does not appear from Leach's data that the Kachins give evidence of intending their statements about these spirits in any sense other than a literal one, the conclusion seems odd. I think the interpretation which I offer here might do better justice to the facts.

^{page 101 note 1} A very clear exposition of this and other features of model-building in the sciences is to be found in Hesse, Mary, Models and Analogies in Science, London 1963.Google Scholar

^{page 102 note 1} For an application of this type of interpretation to yet other features of religious cult and belief in more particular ethnographic contexts, see my ‘Kalabari World View ’, Africa, July 1962. Also ‘The High God in West Africa: a Reply to Father O'Connell’, Man, September 1962.