2 . Gunnar Landtman. 



not observe that he in any way expressed himself witli less fluency than usual. If I 

 enjoined him to relate the legend in its entirety he as a rule expanded the same with 

 equal ease. 



One singnlarity common to many of my informants was their disposition at the 

 beginning of a story to so to speak fumble about with the subject for a good while 

 before coming to the plot. It sometimes happened, that when they began a story they 

 comprised the whole argument into a few broken sentences, absolutely unintelligible, if 

 the legend was not familiär to one before. In other cases the same uncertainty ex- 

 pressed itself in a long drawn out introduction which for instance included a description 

 of the people engaged in their every day occupations, first in the home, and then in 

 the plantations över and över again, till at length the action began. Once however the 

 story was properly started it was pursued with good observance to coherency. Repeti- 

 tions came about often, when certain épisodes reoccurred in the same tale and all the 

 détails were delineated each time with the same fullness. 



The folktales served as a good guidance for my other research work among the 

 natives. In the beginning I devoted a good deal of time to the so called genealogical 

 method of research, but at least in the case of the Kiwais, who live in very populous 

 communities, this method did not prove very satisfactory, as it demanded much time 

 and hardly lead to any other results than an insight into the system of kinship of the 

 natives. The généalogies, in the recording and analysing of which the method consists, 

 could not either in spite of every effort be made as complète or exact as one would 

 have wished. One difflculty in my work among the natives was to avoid secret or 

 otherwise forbidden subjects in order not to frighten my informants before they had 

 accustomed themselves to me. For this purpose I used to begin with asking them to 

 relate folktales which they themselves should choose. This method, which I låter fol- 

 lowed in every place I visited, could be used so much the better as the natives possessed 

 an inexhaustible store of folklore of that description, and imparted legends both better 

 and more readily than anything eise. Episodes which arose in the legends served as 

 stepping stones to other questions. Even certain legends were secret, for instance those 

 concerning the great cérémonies, but such the informants themselves avoided in the 

 beginning. 



The reader will find much in the contents of the legends which is illogical and 

 inconséquent. I have however purposely omitted in this respect to révise the text in 

 any way. For it was part and parcel of the natives" way of narrating that they did 

 not trouble themselves about any logical deficiency in the tales. The legend numbered 

 62 thus speaks of a dumb woman who yet talks. The fitumu, to give another instance, 

 is the headless ghost of a man whose head has been eut off, but no. 134 deals with 



Tom. XL vu. 



