THE writing of this book has been a harder task than I anticipated when I began it in 1938. It has taught me that the hardest part of an anthropologist's work begins after he leaves the field. In the field he is engrossed in concrete human activities. It is true that he can never feel himself completely at one with the people he is studying, however gifted he may be, linguistically or psychologically. He may make some real friends among his hosts; but he can never adopt their cultural values. If he did, he would lose that detachment without which anything he wrote would be of no scientific value. All the same, he is carried by the living stream of social life. He hardly has to stop and think. Thus the crucial scientific task begins when he starts to write up his field material. It is not merely a question of putting his observations on record. Writing an anthropological monograph is itself an instrument of research, and perhaps the most significant instrument of research in the anthropologist's armoury. It involves breaking up the vivid, kaleidoscopic reality of human action, thought, and emotion which lives in the anthropologist's note-books and memory, and creating out of the pieces a coherent representation of a society, in terms of the general principles of organization and motivation that regulate behaviour in it. It is a task that cannot be done without the help of theory.
The task would be easier if social anthropology had a well-founded body of principles to which every new observation in the field could be related. This is not so; and though we all build on the shoulders of our predecessors, working out our problems with their hypotheses and findings implicitly or explicitly in mind, yet every new piece of field-work is unique in some respects, and every new monograph raises unique problems of analysis and exposition. There is inevitably, therefore, a large personal element in an anthropological monograph.
One of my difficulties lay in the complexity of the material I have tried to present. It is commonly accepted that primitive societies are much simpler than ours, both in their cultural make-up and in their social and economic structure. But even a very simple cultural heritage or social organization may be composed of a large number of strands and elements whose interrelations make up a very elaborate pattern. Simplicity in this context is much more a value concept than a scientific category. In social science it is partly a question of how closely and with what discipline one examines the data. I should have found it much easier to write this book at the end of my first tour among the Tallensi, but it would have been inaccurate in some points and defective in many. I did not know enough about the External Bayar cult of the Hill Talis, for example, at the end of my first tour, to understand its significance in the social structure. Indeed, this and many other matters only became clear to me in the course of writing this book.