96 THE EVOLUTION OF MALAY SPELLING. 



" while it shakes my belief in their accuracy, has not been able to 

 "destroy it; and so much the less because Mr. van de Wall's 

 11 presentation of the subject is itself not quite clear. He s^js 

 " (73, 399) ' the first result of the suffixes kan, i, an, nya, 1-u, la/t, 

 k - tah, kah, on roots which end in an open syllable is, that they leng- 

 " then the vowel of that syllable and cause the original long 



" vowel of the root to drop out : the suffixes an and 



"t, since they begin with a vowel, cause the same result in 

 '.' words which end in a closed syllable, and in that case the final 

 " consonant of the root becomes the initial letter of the suffix 



" with the corresponding vowel while the other 



" suffixes leave such words unchanged ; but the 



" change of position of the lengthening letter to or its appear- 

 "ance in the penultimate of root words (read, of derived words ? 

 " or in the last syllable of root words) has in most words no in - 

 " liuence upon the accent, or at least very little ; that is to say, 

 " the accent is in such cases not inherent in the long vowel.' 



" Here first of all the question arises : is the change of posi- 

 " tion of the lengthening letter a mere graphic phenomenon, 

 " does it only exist in the Malayo- Arabic character and the trans- 

 " literations thereof, or does the same change in the length of 

 " the vowels take place in the pronunciation ? If this is maintain- 

 k * ed, I must then further ask how such a rule can have arisen in 

 " the written character, a character which so to speak does not 

 " belong to the language, and if such were the case might be 

 kk expected to have preserved in this respect the traces of a long 

 " obsolete condition of the language or perhaps of some kindred 

 " dialect, but which, borrowed from an entirely foreign language 

 " and probably first applied to the Malay in comparatively recent 

 " times, must be reckoned as rendering the native pronunciation in 

 " common use as accurately as the foreign characters will allow ? 

 " I could understand that the retention of the original spelling of 

 " a root ending in a consonant when followed by a suffix beginning 

 " with a consonant, might arise from an idea of producing legi- 

 " bility, so as not to entirely deprive the word of vowel signs, 



"and that one might therefore write for instance -x;j\_xu 

 " although perhaps (according to Werndly ) jC;J ja* might better 



