attempts to restore the orthographical principles of three centuries ago, as (a) we have to make use of the Roman alphabet which is more suitable for expressing Malay vocalism and more in accordance with actual requirements of conversation than the Arabic; (b) the pronunciation of that time is very imperfectly known to us; (c) it has everywhere changed a good deal. So a mean will have to be sought between two extremes. The language of Malacca (Riau-Johor) having played a prominent part in fixing the κοινὴ διάλεκτος of the golden age, and so having left considerable traces in what has remained of the old uniformity of civilized speech, may be given more authority than most of its sister dialects, but these last nevertheless have to be taken very seriously into account. The local diversity of vocalism must keep us from marking too fine distinctions of sound and from fixing in script nuances which in a considerable part of Malaya are not observed.’
To put this general principle into practice:
(a) How is the Malay indeterminate short vowel to be represented? Some old writers used a: bassar great, marampas seize; Crawfurd affected â—bârchârai divorced; Marsden and older Dutch scholars e, besar. The missionary press in Singapore wavered between omitting it—bsar—and inserting ’ or ′—b’sar, b′sar. Swettenham, van Wijk and modern continental scholars generally have preferred ĕ—běsar. Now a and e represent its sound neither to the Malay nor to the foreigner; and e will be confused with e in words like beta and besan. Use no symbol and an agglomeration of consonants—kbsaran—results. Use ’ or ′ and ĕngkau becomes ’ngkau and kěĕnam becomes k’’nam! Obviously ĕ best represents its sound and is no harder to write than a dotted i or a French accent or the German diaeresis.
(b) How are ڠ ng as in دڠن dengan, اورڠ orang, and ڽ ny as in ڽاموق nyamok, باڽق banyak to be written? The scientific
