[CAMPBELL] ORIGIN OF THE HAIDAHS 93 



Crawford supposes the Hindoo migration to the Malay archipelago 

 to have begun in the thirteenth century, and it is supposed that this and 

 tlie subsequent Mahometan invasions caused large displacements of popu- 

 lation. That the expulsion of the Haidahs was posterior to the rule of 

 the Hindoos seems evident, from their possessing the Sanscrit surya, as 

 the name of " the sun," in the form izoore. It is also possible that the 

 Haidah kung, the moon, is a form of the Sanscrit chandra abbreviated. 

 Malay domination has stamped itself upon the language in the word for 

 man, orang, olamj, which is the Malay's peculiar property, and which the 

 Haidah but faintly disguises in eetling and ihiiiKja. Other Malay terms, 

 such as perampuan, woman ; kapala, head ; viata, eye ; telinga, ear ; 

 tangan, lima, hand; nana, house; biwii, tanah, earth; api, fire; baik, 

 good ; jahat, bad, etc., are conspicuous by their absence, while the Papuan 

 and Australian forms are exceedingly numerous. The absence of labials 

 in Haidah, the place of which is taken generally by the sound of iv, some- 

 times by an aspirate, and but rarely by the liquid m, exhibits phonetic 

 decay not uncommon in American dialects, and renders perplexing, at 

 first sight, the identity of compared words. Another source of difficulty 

 is the combination tl, which is not characteristic only of Aztec speech. 

 In most cases it appears to be an expedient for an original I or r, as in 

 eeiling for orang. This is a mere matter of dialectic variation, as appears 

 from a comparison of the various forms of Caucasian speech. Thus, in 

 Lesghian, the Avar word for night is rahle, but in Audi it is retlo. The 

 sun again is beri in Akush, and mitli in Audi. The Nicaraguan dialect 

 of the Mexican reveals the same equivalency, the Aztec Nahuatl being 

 its Nahuar. 



What stamps the Haidah as a Melanesian language is its grammatical 

 <;onstruction, in which it differs entirely from the Malay and the Poly- 

 nesian proper. These latter are preposing languages, which does not 

 simpl}' mean that they make use of prepositions, but that they also place 

 the governing woi'd before its genitive, the temporal index before the 

 verbal root, and, generally speaking, the abstract before the concrete as 

 in Semitic and Indo-European speech. The Melanesian languages, in 

 general terms, do the very opposite, and are thus postpositional, in all of 

 which respects the Haidah agrees with them. A comparison of Mr. 

 Harrison's Haidah grammar with Threlkeld's Australian one, does not 

 indeed reveal identity of structure, which would be remarkable, but it 

 exhibits so many and such striking points of resemblance as to show that 

 the two languages belong to one and the same family. In the Malay 

 archipelago the presence of the same syntactical order may easily be 

 detected, even within the compass of a brief vocabular}-. Xow, the 

 Malay calls the middle of the night tangah inalam, in which tangah is 

 middle and malam is night ; but the native of Teor terms a finger-nail 

 ■limin-kukin, in which limin is hand and kukin is nail. "While both seem 



