Published literature on the subject of the Shilluk is considerable and there are also many interesting accounts of specific subjects buried in the files of the Administration. Nowhere, however, is there a complete account of all aspects of Shilluk life. Moreover, there has been a tendency to concentrate on certain aspects, notably the Divine Kingship, to the exclusion of others. In 1941 Mr. M. E. C. Pumphrey and I published two articles: the first a general summary account of the Shilluk tribe based on Mr. Pumphrey's observations and wide experience; the second a more detailed analysis of one Shilluk settlement, the result of fieldwork carried out in 1937. In 1944 I had the privilege of attending the installation of reth Anei Kur and, together with Mr. W. P. G. Thomson, published a full account. Subsequently Mr. Thomson published his ‘Further Notes on the Death of a Shilluk Reth’, after the death of reth Anei in 1943. This, with Professor Evans-Pritchard's Frazer Lecture, The Divine Kingship of the Shilluk of the Nilotic Sudan, represents the total of information published on the subject of the Shilluk in recent years.

Certain features of Cewa witch beliefs make them well suited to sociological analysis. As this type of analysis has been somewhat neglected by students of witch beliefs, Part I of this paper will be devoted to a brief review of some of the literature (mainly on Bantu) and to an attempt to formulate a serviceable hypothesis. Part II will summarize information about the Cewa essential to an understanding of Part III, in which the hypothesis will be applied to Cewa material. Part III will appear in a subsequent number of Africa.

The Ethiopic syllabary employed for writing the classical Ge'ez and also, with certain modifications, the contemporary South Semitic vernaculars of East Africa, was formed by super-imposing a system of auxiliary vowel-marks upon a basic consonantal alphabet; this alphabet occurs, alongside of the syllabic script, in the Old Ethiopic inscriptions of the Axumite Kingdom in the fourth century of our era, and is a derivative of the Sabaeo-Minaean or Old South Arabic script found in the monuments of the south-west Arabian kingdoms. But although the Ethiopic syllabary is thus genetically connected with the other main branches of the Semitic alphabet, the traditional order of the signs, in which the consonantal component and the accompanying vowel are the primary and secondary determining factors respectively, does not agree with that of any Semitic alphabet hitherto known. There is no old or reliable native tradition as to the reason underlying the order of the signs; no help is to be had from numerical signs, which elsewhere, as will shortly appear, afford valuable testimony to the order of the letters; for Ethiopic borrowed Greek alphabetic signs for this purpose, while the South Arabian inscriptions used single strokes for the units, and for higher denominations the initial letters of the native words for five, ten, hundred, &c. The mnemonic word-groups reconstructed by Bauer and others are open to objection on grounds of language and sense. Other external criteria yield only tentative and inconclusive results, and the subject has accordingly remained one of speculation and controversy.

Certain sampling problems which arose in the design of the Agricultural Survey of Basutoland in 1949-50 were solved by the use of the native tax registers, in conjunction with the rather scanty information provided by the last population census in 1946. As the situation was one which is likely to be found in a number of African Territories, it may be of interest to describe it and to discuss the methods by which our difficulties were overcome.

Swahili is, by origin, a Bantu language which was strongly influenced by Arabic. In the course of time Swahili has borrowed words from a variety of languages. During the last thirty or forty years it is not surprising to find that Swahili has continued to enrich its vocabulary, more particularly from English.

In countries where conditions favour the preservation of manuscripts and books, the survival value of much of the literature may not be directly related to its popularity. Unpopular works may remain neglected for long periods and then regain favour; but in the conditions which have prevailed in East Africa, only those literary works receiving the continued favour of the people have had any chance of survival at all. There have been no library stacks, no dry cupboards or shelves for storing manuscripts, but only the white ant and the damp. The love of the Swahili people for a good poem has been the means of retaining much valuable work that would otherwise have been lost. Even so, much more has been lost to posterity than has been preserved.

In Western society children grow up with certain elementary ideas of the costs and values of economic goods. As they mature, their personal contacts and their reading still further impress them with the importance of effort, saving, and sacrifice in relation to reward. Their concepts of reward are enlarged and these concepts become clothed in a great variety of specific forms, which become their standards of living, and serve as incentives to further effort. That is to say, children learn, consciously or unconsciously, from the beginning, to value a large number of economic goods which they recognize as costly and needing to be cared for. They learn also that economic rewards come only with striving. And because they have experience that the opportunities which make rewards possible are precious and somewhat limited they come to prize them, conserve them, and make use of them when they are offered.

Parmi les vœux adoptés à l'unanimité et sans discussion par le Deuxième Congres de toponymie et d'anthroponymie, tenu à Paris du 15 au 17 juillet 1947, il en est un auquel il faut donner la plus large diffusion: ‘Le Congrés . . 'met le vceu que les noms propres de lieux retenus pour les cartes d'Afrique Occidentale Française et d'Afrique Équatoriale soient ceux qu'emploient les Indigènes habitant les lieux, dans leur propre langue, et non les vocables utilisés par les populations voisines, qu'elles emploient ou non une langue vehiculaire . . .’ Ce vœu doit être complété par cet autre qui ne concerne plus les toponymes mais les noms ethniques: que pour désigner une ethnie on n'emploie que le vocable utilisé dans la langue de ce peuple pour se désigner lui-même, à l'exclusion des vocables utilisés par les populations voisines, qu'elles emploient ou non une langue véhiculaire. Salomon Reinach disait que les noms ethniques étaient la peste de l'anthropologie, cette affirmation reste vraie pour l'ethnologie.