678 SYSTEMS OF CONSANGUINITY AND AFFINITY 



REMARKS OF MR. FISON ON THE TONGAN SYSTEM. 



From these relationships it s';ems that the Tongau system differs materially from the Fijian as to 

 " cousinage" — all victavaleni, veidavolani, and veiraivani, excepting the first generation, being 

 tautehina {= veitacini) in the Tongan. 



I have spelt all words after the Tongan manner, making a few exceptions for the sake of clearness, 

 but departing slightly from the Tongan system of spelling only where that system is manifestly 

 faulty. For instance, the Tongan language has the sound of p as well as that of b (the English h, 

 not the Fijian, which is mb) ; but the fcamers of the alphabet have rejected p altogether, making b 

 do double duty. I have, therefore, spelt "matapule," " takape," &c., with the p, though they would 

 be spelt by a Tongan " matabule," " takabe," &c. They are sounded as I have spelt them. 



Again, the framers of the Tongan alphabet have used aa very often where a only is required ; and 

 this is disastrous, because there are many words which require the aa, as ma'a — words in which 

 there is a break. 



Speaking of these breaks, I am inclined to think that they represent missing letters — letters which 

 were formerly sounded in the word. What makes me almost sure of this, is the singular fact, that 

 here in Fiji, two dialects, not differing materially from the Bauan, drop, one the k, and the other 

 the t, in every word wherein these letters occur, making a break in the sound of the word where the 

 missing letter has fallen. Thus, the Cakandrove people say " 'ata'ata," where a Bauan says "kata- 

 kata ;" and the people on a part of the coast of Navitilevu pronounce the same word " ka'aka'a." 

 I have found this difference existing between two i-slands not ten miles apart ; nay, even between two 

 towns on the same island. Thus, on the island of Vanua Balavu, the Lomaloma folks say " kata- 

 kata," but the Mualevu folks, '"ata'ata." The Ovalau islanders say " e tini na tamata" = ten men, 

 while the people of the neighboring island, Naigani, say "e 'ini na 'ama'a." The break is distinctly 

 heard, even when it occurs at the beginning of the word. Now there is no physical difBcuIty to 

 account for this singular fact, no physical difiSculty such as the Northu»ibrians have in pronouncing 

 the letter r. " Au sa la'o 'i na 'oro," said a Cakandrove woman whom I met in the Bau district 

 (= " I am going to the town"). " Vosa vaka bau mada ga," said I (= " speak Bauan") ; where- 

 upon she said at once, with a laugh, " Au sa lako ki na koro." 



I can, therefore, look upon this letter dropping only as mere caprice, that is as to the origin of it. 

 It may have taken its rise in the determination of some ruling chief never to pronounce a letter 

 which was the first or predominant letter of some hated enemy's name ; and he may have commanded 

 his people to follow his example, or they may have followed his example of their own accord. It 

 may have taken its rise, however, in mere affectation, just as I remember the gobbling sound of the 

 letter "r" took its rise among foolish young men of the Lord Dundreary stamp. 



The sound of the Tongan vowels is that of the Fijian, and the remarks which I made about the 

 so-called Fijian diphthongs will apply to the Tongan also. 



G is pronounced ng as in Fijian ; but 



B is never pronounced mb. 



I subjoin a list of the words employed, with the accented syllables and the long vowels marked. 

 I have separated certain syllables when it has seemed necessary to do so for the sake of clearness. 

 The short vowels are unmarked. 



