502 



The division into chapters and sections which I have 

 adopted needs also a word or so of explanation. By adopting 

 a fairly systematic division — Geography, Sociology, Economics, 

 Magico-religious matt-ers, etc. — I have, of course, deviated 

 from a purely topical presentation of facts. I shall undoubt- 

 edly be taken to task for having thus "put some theory into 

 the facts," and, in the first place, for having swerved from the 

 native forms of tliinking and abandoned the native classifica- 

 tion . 



A thorough reply to such an objection would require a 

 long philosophical digression, for which there is no room here. 

 I wish, however, to emphasize the fact that I have always tried 

 to leave all preconceived ideas on one side when working among 

 the natives, and to classify and register facts according to their 

 essentials as I saw them then and there. In discussing facts, 

 which may be called religious or magical, according to the 

 student's point of view and definition, I have followed Dr. 

 Marett's excellent advice and hints, given in the '"'Notes and 

 Queries," and, in accordance therewith, I have called them 

 throughout by the impartial term Magico-religious. I have 

 tried to preserve the same attitude in all the various subjects 

 investigated. 



Again, I have always endeavoured to preserve as far as 

 possible the natives' own way of looking at things, and when- 

 ever I was able to obtain a native classification of facts — as, for 

 instance, that implied in the term Gora (chap, iii., sec. 5) — I 

 was careful to record it and to state it as exactl}?- as possible. 



In the spelling of native names I have adopted the system 

 recommended by the ''Notes and Queries," which practically 

 consists in the use of the consonants with their English sounds 

 and the vowels with their Italian values. I found this system 

 entirely sufficient with Mailu. The only sign I found necessary 

 to introduce refers to the distinction between the diphthongal 

 sounding of two vowels and their separate sounding. Wherever 

 the vowels are simply juxtaposed they are to be sounded 

 dipthongally. Wherever the two vowels retain their separate 

 values I have indicated this by putting an apostrophe between 

 them, but it is important to note that the apostrophe so placed 

 does not imply any pause or any specific sound, similar to the 

 Arabic am. Thus in the word Bo'i the apostrophe between o 

 and I means that the vowels have to be pronounced as in the 

 Italian poi, and not as in the English hoy. I have also put an 

 accent upon nearly every word, except those in which I failed 

 to record it through haste in taking the notes. On some com- 

 pound words I have put a double accent. As a matter of 

 fact, the natives join the component parts and pronounce them 



