Chapter XI

) must be sufficient and we shall confine ourselves to their tapwana. The word gwa'u or ga'u means 'mist' or 'fog'; verbally used with the meaning 'to make mist' 'to befog,' it has always the form ga'u. In the main parts of some of the formulæ of this class, this phonetically very expressive word is used with very great sound effect. For example in the giyorokaywa spell No. 1, the key-words are aga'u ('I befog'), aga'usulu ('I befog, lead astray'); aga'uboda ('I befog, shut off'). Spoken, at the beginning of the tapwana slowly and sonorously, and then quickly and insistently these words produce a really 'magical' effect--that is as far as the hearers' subjective impressions are concerned. Even more impressive and onomatopoetic is the phrase used as key-expression in the Giyotanawa No. 2:

Ga'u, yaga'u, yagaga'u, yaga'u, bode, bodegu!

This sentence, giving the vowels a full Italian value, such as they receive in the Melanesian pronunciation, does certainly have an impressive ring; fittingly enough, because this is the dramatic spell, uttered into the wind in the sinking waga, the final effort of magic to blind and mislead the mulukwausi. The causative prefix ya- is used here with a nominal expression yaga'u which has been translated 'gathering mist'; the reduplicated one yagaga'u I have rendered by 'encircling mist.' It can be seen from this example how feebly the equivalents can be given of the magical phrases in which so much is expressed by phonetic or onomatopoetic means.

The other spells have much less inspired key-words. Giyotanawa No. 1 uses the word atumboda, translated 'I press,' 'I close down,' which literally renders the meanings of the verbs tum, 'to press,' and boda, 'to close.' The Giyorokaywa No. 2 has the somewhat archaic key-words spoken in a couple: 'apeyra yauredi,' 'I arise,' 'I escape' and the grammatically irregular expression suluya, 'to lead astray.'

The main part of the Kaytaria spell, by which the benevolent fish is summoned to the rescue of the drowning party has the key-phrase 'bigabaygu suyusayu: the suyusayu fish shall lift me up.' This expression is noteworthy: even in this spell, which might be regarded as an invocation of the helpful animal, it is not addressed in the second person. The result is verbally anticipated, proving that the spell is to act through the direct force of the words and not as an appeal to the animal.

XII

With this, the survey of linguistic samples from various spells is closed, and we can briefly summarise our results. The belief in the efficiency of a formula results in various peculiarities of the language in which it is couched, both as regards meaning and sound. The native is deeply convinced of this mysterious, intrinsic power of certain words; words which are believed to have their virtue in their own right, so to speak; having come into existence from primeval times and exercising their influence directly.

To start first with the meaning of the magical expressions, we have seen that in this respect they are plain and direct enough. Most of the key-words simply state the magical action, for example when in one of the spells the key-word napuwoye, 'I impart magical virtue (of speed),' or in another the key-words 'to paint red in a festive manner, to wreathe in a festive manner,' simply describe what the magician is doing. Much more often the principal expressions, that is the initial words and the key-words, of a spell refer to its aim, as for instance, when we find words and phrases denoting 'speed' in canoe magic; or, in Kula magic, designations for 'success' 'abundant haul,' 'excitement,' 'beauty.' Still more often the aim of magic is stated in a metaphorical manner, by similes and double meanings. In other parts of the spell, where the magical meaning is imprisoned not so much in single words and expressions, as in explicit phraseology and long periods, we found that the predominant features are: lists of ancestral names; invocations of ancestral spirits; mythological allusions; similes and exaggerations; depreciating contrasts between the companions and the reciter--most of them expressing an anticipation of the favourable results aimed at in the spell. Again, certain parts of the spell contain systematic, meticulous enumerations, the reciter going over the parts of a canoe one by one; the successive stages of a journey; the various Kula goods and valuables; the parts of the human head; the numerous places from which the flying witches are believed to come. Such enumerations as a rule strive at an almost pedantic completeness.

Passing to the phonetic characteristics, we saw that a word will often be used in a shape quite different from those in which it is used in ordinary speech; that it will show notable changes in form and sound. Such phonetic peculiarities are most conspicuous in the main words, that is in the key-words and initial words. They are sometimes truncated, more often provided with additions, such as symmetrical or antithetic affixes; formatives added for the sake of sound. By these means there are produced effects of rhythm, alliteration and rhyme, often heightened and accentuated by actual vocal accent. We found play on words by symmetrical couples of sounds, with antithetic meaning like mo- and vi-, or mwana- and vina-, both couples signifying 'male' and 'female' respectively; or -mugwa (ancient) and -va'u (new); or ma- (hither) and wa- (thither), etc., etc. Especially we found the prefix bo-, carrying the meaning of ritual or tabooed, with derivation from bomala; or with the meaning 'red,' 'festive' in its derivation from bu'a (areca-nut); onomatopoetic sounds such as sididi or saydidi, tatata, numsa, in imitation of speed noises, of the wailing of wind, rustling of sail, swish of pandanus leaves; tududu, in imitation of the thunder claps; and the rhythmical, expressive, though perhaps not directly onomatopoetic, sentence:

Ga'u, yaga'u, yagaga'u, yaga'u, bode, bodegu.

XIII

If we now turn to the substances used in the magical rites, as means of ritual transference of the spell, we find in canoe magic, dried lalang grass, dried banana leaf, dried pandanus leaf, all used in the magic of lightness. A stale potato is employed to carry away the heaviness of the canoe; although on another occasion heaviness is thrown away with a bunch of lalang grass. The leaves of two or three shrubs and weeds, which as a rule the natives take to dry their skin after bathing, are used for magical cleansing of a canoe body, and a stick and a torch serve in other rites of exorcism. In the rite associated with the blackening of a canoe, charred remains of several light substances such as lalang grass, the nest of a small, swift bird, the wings of a bat, coco-nut husk and the twigs of an extremely light mimosa tree are employed.

It is easy to see that, not less than the words, the substances here used are associated with the aim of the magic, that is, with lightness, with swiftness and with flying.

In the magic of the Kula we find betel-nut, crushed with lime in a mortar, used to redden the tip of the canoe. Betel-nut is also given to a partner, after it has been charmed over with a seducing spell. Aromatic mint, boiled in coco-nut oil and ginger root are also used in the mwasila. The conch-shell, and the cosmetic ingredients, charmed over on Sarubwoyna beach are really part of the outfit, and so is the lilava bundle. All the substances used in this magic are associated either with beauty and attractiveness (betel-nut, cosmetics, the mint plant) or with excitement (conch-shell, chewed betel-nut). Here therefore, it is not with the final aim--which is the obtaining of valuables--that the magic is concerned, but with the intermediate one, that is that of being agreeable to one's partner, of putting him into a state of excitement about the Kula.

XIV

I wish to close this chapter by adducing a few texts of native information. In the previous chapters, several statements and narratives have been put into the natives' mouths and given in quotations. I wish now to show some of the actual linguistic data from which such quotations have been derived. Numerous utterances of the natives were taken down by me as they were spoken. Whenever there was a native expression covering a point of crucial importance, or a characteristic thought, or one neatly formulated, or else one especially hazy and opalescent in meaning--I noted them down in quick handwriting as they were spoken. A number of such texts, apart from their linguistic importance, will serve as documents embodying the native ideas without any foreign admixture, and it will also show the long way which lies between the crude native statement and its explicit, ethnographic presentation. For what strikes us at first sight most forcibly in these texts is their extreme bareness, the scantiness of information which they appear to contain. Couched in a condensed, disjointed, one might say telegraphic style, they seem to lack almost everything which could throw light on the subject of our study. For they lack concatenation of ideas, and they contain few concrete details, and few really apt generalisations. It must be remembered, however, that, whatever might be the importance of such texts, they are not the only source of ethnographic information, not even the most important one. The observer has to read them in the context of tribal life. Many of the customs of behaviour, of the sociological data, which are barely mentioned in the texts, have become familiar to the Ethnographer through personal observation and the direct study of the objective manifestations and of data referring to their social constitution (compare the observations on Method in the Introduction). On the other hand, a better knowledge of and acquaintance with the means of linguistic expression makes the language itself much more significant to one who not only knows how it is used but uses it himself. After all, if natives could furnish us with correct, explicit and consistent accounts of their tribal organisation, customs and ideas, there would be no difficulty in ethnographic work. Unfortunately, the native can neither get outside his tribal atmospheres and see it objectively, nor if he could, would he have intellectual and linguistic means sufficient to express it. And so the Ethnographer has to collect objective data, such as maps, plans, genealogies, lists of possessions, accounts of inheritance, censuses of village communities. He has to study the behaviour of the native, to talk with him under all sorts of conditions, and to write down his words. And then, from all these diverse data, to construct his synthesis, the picture of a community and of the individuals in it. But I have dwelt on these aspects of method already in the Introduction and here I want only to exemplify them with regard to the linguistic material directly representing some of the natives' thoughts on ethnographic subjects.

XV

I shall give here first a text on the subject of the priority in sailing, which as described in