38 Journal of the Asiatic Society of Bengal. |February, 1911. 
and the difficulty is still greater when we have to do with 
strange tongues. 
To return to the Mundari vowels it would have been 
impossible in a work like the Linguistic Survey to give an 
exposition of the phonetic system on which the marking of the 
various sounds has been based. The ear is often a very unsafe 
guide, and the marking of vowels in phonetic books is 
therefore based on an analysis of ‘the various positions of the 
ngue. I cannot do better than to quote Mr. Sweet! in order 
to explain this. He says :-- 
‘As each new position of the tongue produces a new 
vowel, and as the positions are infinite, it follows that the 
number of possible vowel-sounds is infinite. It becomes neces- 
sary, therefore, to select certain definite positions as fixed points 
whence to measure the intermediate positions. 
The movements of the tongue may be distinguished gene- 
rally as horizontal and vertical—backwards and forwards, up- 
wards and downwards. The horizontal movements produce 
and a@ in man, the front of the tongue is raised towards the 
front of the palate, so that the main body of the tongue slopes 
down from the front of the mouth backwards. There is a third 
class of ‘ mixed’ (gutturo-palatal) vowels such as the ¢ in err, 
where the whole tongue is allowed to sink with its neutral 
flattened shape, in which neither back nor front articulation 
predominates. 
ally accompanied by lowering and raising of the jaw, produce 
the a 1n man | it is lowered as much as possible. From among 
the infinite degrees of height three are selected : (1) ‘ high’ 
[as in fill), (2) ‘mid’ [as in men], (3) ‘low’ [as in man]. 
hese distinctions apply equally to back and mixed vowels, so 
we have altogether nine cardinal vowel-positions :— 
high back high mixed high front. 
mid back mid mixed mid front. 
low back low mixed low front.’ 
FES ag Le ei i 
| A Primer of Phonetics. Second edition. Oxford, 1902. pp. 13 and f. 
