216 THE LANGUAGES OF THE INDIAN ARCHIPELAGO* 
sufficiently interested in the subject to read the essay or use 
the dictionary, will take the trouble of attending to the few 
simple rules on which correct enunciation depends. In the 
comparative philology of the Archipelago, in which we draw 
results and adduce illustrations from numerous languages and 
dialects, with many of which the reader may be totally unac¬ 
quainted, these accentual and prosodial marks cannot be dis¬ 
pensed with. In geographical and ethnological essays they 
are also necessary. 
Our scheme, like all those which adhere to the Homan 
vowel symbols, is one of accommodation and compromise. 
We deprecate the assumption by individuals of the power of 
inventing new letters. All such invasions of our time honoured 
A. B. C. must fail of success, until philology becomes at once 
more scientific and more popular, and national prejudice les* 
strong, when we may hope to see a phonetic alphabet adopted 
in Europe.* Those only who have compared a considerable 
group of eastern languages with each other, and with the more 
current European languages, in their vocalic sounds, can 
appreciate the labour and difficulty of correctly ascertaining 
and classifying these sounds, and the hopelessness of ex¬ 
pressing them by the letters of the Roman alphabet. So 
long as we must do so descriptively instead of symbolically, 
our progress in this rudimentary portion of philology will 
be unsatisfactory. Let it not be thought that we advocate 
a too microscopic investigation of the elements of language, 
or seek to trace natural laws beyond the limits where human 
caprice shuts them out. We have a deep persuasion that 
we cannot err on this side, and that the only reason why there 
does not yet exist a science of language is, that very few 
of its cultivators have applied to it the keen observation 
and exact discrimination of facts, which have furnished the 
basis, and the sound and severe methods of generalization 
which have built up the fabric, of the natural and physical 
sciences. It is only by a minute analysis and comparison of 
the elementary sounds and vocal laws of each language in a 
group like that of the Archipelago, that we can understand 
those numerous and often complicated or obscure metamor- 
t We see a phonetic newspaper is advertised in England. The new letters 
to be used are necessary, but their shape in several instances does not harmo¬ 
nize with the general character of the Roman letters. Mr Hale’* innovation* 
are open to the same objection, but he does not recommend them for adoption, 
and when a philologist has recourse to new letters for his own purposes it la 
perhBps best that those which he engrafts on the Roman alphabet should be 
borrowed from the Greek, or any other well-known alphabet which containi the 
•ymbol* wanted. 
