businesslike performances. In calling a magical action ‘ceremonial’ we imply that it was done with a big public attendance; under the observance of definite rules of behaviour by the spectators as well as by the performer, such as general silence, reverent attention to what is being done, with at least a show of some interest. Now if, in the middle of some work, a man quickly performs an action whilst others talk and laugh and leave him entirely on one side, this gives a definite sociological stamp to the magical actions, and does not allow us to use the term ‘ceremonial,’ as the distinguishing mark of the magical acts. Some of them, it is true, do have this character. For instance, the initial rite with which the kaloma fishing begins, requires the assistance of the whole fleet, and a definite type of behaviour on the part of the crews, while the magician officiates for all of them, but with their assistance, in the complex evolutions of the fleet. Similar rites are to be found in two or three systems of fishing magic, and in several rites of the garden magic of certain villages. In fact, the initial rite of garden magic is everywhere connected with a ceremonial performance. The garden rite, associated with the ceremonial offering of food to spirits, and attended by a body of villagers, a scene of which is shown on Plate LIX, has been elsewhere described.3 One or two rites in war magic imply the active assistance of large numbers of men, and take the form of big ceremonies. Thus we see that magical rites may or may not be ceremonial, but that the ceremonial is by no means an outstanding or universal feature of Trobriand magic.
X
We found that taboos are associated with magic, in so far as it is the magician who has to observe them. There are, however, certain forms of restrictions or prohibitions, set up for special purposes, and associated with magic in a somewhat different form. Thus, in an institution called kaytubutabu we find a ban made on the consumption of coco-nuts and betel-nuts, associated with a specific magic to make them grow. There is also a protective taboo, used to prevent the theft of ripening fruits or nuts, too far away from the village to be watched. In these cases a small parcel of medicated substance is placed on the tree or near it, on a small stick. The magic spoken over such a substance is a ‘conditional curse,’ to use the excellent term introduced by Professor Westermarck. The conditional curse would fall upon anyone who would touch the fruits of that tree, and would bring upon him one form of disease or another. This is the only form of magic, in which the personal agency is invoked, for in some of these spells, the tokway (wood-sprite) is invited to take up his abode on the kaytapaku, that is the stick, with the substance on it, and to guard the fruit. Some such small divergencies from the general trend of native belief are always to be found. Sometimes they contain important clues, and a deeper insight into the facts, sometimes they mean nothing, and only emphasise the fact, that it is not possible to find absolute consistency in human belief. Only a deeper analysis, and a comparative study of similar phenomena can decide which is the case.
XI
In order to complete the survey of all the characteristics of magic, I shall rapidly mention here the economic aspect of the position of magician, although the data referring to it have already been given, scattered through the previous chapters. I have spoken of the matrilineal inheritance of magic, and of the deviations from it which consist in inheritance from father to son, and in the transmission of magic by purchase (Chapter II, Division VI, and Chapter VI, Division VI under (5)). This latter transaction may take place under two names, which really cover two essentially different operations; the pokala or payment to a maternal kinsman from whom one is going to obtain the magic, and the laga, which is the purchase of magic from