THE ISLAND OF BALI, 
123 
conclusions of Humboldt must be introduced by the fact that 
pure Sanskrit writings are still found with the priests on 
Ball. 
Kawi is explained by Humboldt to be (( poetical language” 
(Kawi ei a poet,” kawya c( a poem.”) With this explanation 
that of the Balinese agrees ; they say that Kaiuin or Kaka - 
win signifies (e to make comparisons” fe to speak in compa¬ 
rison,” This is the mode in which poetry is formed ; com¬ 
parisons are the ornaments and marks of poetry. The 
explanation of the Javanese by Khawi (strong) scarcely 
needs to be mentioned. Khawi is an Arabic word ; first 
known in Java in the Mahomedan era, and in Bali not at all. 
How could the Arabs have given the name to a language 
which they neither produced nor cultivated, but on the con¬ 
trary have destroyed, because it was die prop of Hinduism 
and of all the institutions on Java which the Arabs sought 
to overthrow and cast into oblivion ? The Arabs and their 
followers have succeeded so well, that the Kawi no longer 
exists oyi Java, and Kawi works have nearly disappeared, 
whilst it has been abundantly and carefully preserved on Bali. 
The verb kawin or kakawin has been applied to the words 
which, under that name, are used by the Balinese in m arriage 
songs, while it reminds us of the Malay kawin (to marry ) 
Both words, the Balinese and the Malayan, appear to be 
referrible to the same Sanskrit word. From kawya by the 
suffixing of the Polynesian an, kawyan is formed •, this, by the 
contraction of ya to e commonly (however improperly) used 
in Java, gives kawen , and from this, by a careless pronun¬ 
ciation with the common permutation of e and i, are formed 
kawin and kakawin . This is then at once the Balinese 
word for u poetry” and the Malay for (e to marry,” because 
the marriage songs (hymenaea) form a principal part of 
the festivity, and that which most strikes the ear. Res¬ 
pecting the origin of the Kawi language some new ideas 
must be introduced. The priests do not hold the Kawi 
but the Sanskrit as the sacred language; this language is 
still found on Bali in the vedas, the Brahmandapurana and 
other mystic writings (tutur.) We cannot therefore agree 
with Crawfurd who considered that the Kawi was the lan¬ 
guage of the priests [Crawf. Arch. II, 1 7, IB.] 
The Hindus, and particularly the Hindu-Brahmans who 
came to Java, brought with them the Sanskrit in their sa¬ 
cred writings, and perhaps also a Prakrit dialect. That they 
knew and could speak a Prakrit dialect may be concluded 
from the comparatively late periods of their arrival from 
o 
