A Comparative Grammar of the Dravidian or South-Indian Family of Languages c4: Note on Transliteration
A Comparative Grammar of the Dravidian or South-Indian Family of Languages c4
- Language
- English
Note on Transliteration
c4.1
ALL foreign words, to whatever family of languages they may belong, are represented in this work in Roman characters, for the double purpose of preventing unnecessary expense and trouble, and of facilitating comparison.
c4.2
Long vowels are invariably marked thus,—ā: when no such accent is placed over a vowel, it is intended that it should be pronounced short. *E* and *o*, being invariably long in Sanskrit, are left unaccented in the transliteration of Sanskrit words in works treating of Sanskrit. The Dravidian languages having short *e* and *o*, as well as long, it is to be understood that they are to be pronounced short when unaccented.*
c4.3
All vowels are pronounced in the Continental manner. *ei*, as will be explained, corresponds to the Sanskrit *ai*.
c4.4
The "lingual" or "cerebral" consonants are denoted by a subscribed dot—e.g., *ṭ*, *ḍ*, *ṇ*: the peculiar vocalic *r*, and the surd *l*, of the South Indian languages are denoted in a similar manner—e.g., *ṛ*, *ḷ*: the obscure, inorganic nasal *n* or *m* is represented by *n* with a superscribed line—e.g., *n̄*: the nasal of the guttural row of consonants, ordinarily represented by *ng*, is written *ṅ*; the nasal of the palatal row, ordinarily written *nj* or *ny*, is written *ñ*; and the hard rough *r* is represented by a heavier letter **r**.
c4.5
The dental *d* in Tamil, and the corresponding *t* or *d* in Malayâlam, are pronounced in the middle of a word, or between two vowels, like the English *th* in *than*; and in Telugu, *j* and *ch*, when followed by certain vowels, are pronounced like *dz* and *ts*: but as these are merely peculiarities of pronunciation, and one consonant is not exchanged for another, no change has been made in the characters by which those sounds are represented.
c4.6
I have found it very difficult to determine how the third consonant in Tamil, answering to the Sanskrit *ch*, should be represented. The difficulty is owing to the circumstance that its pronunciation, when doubled, differs considerably from its pronunciation when single. When single, its pronunciation closely resembles that of the Sanskrit ś; when doubled, it is identical with that of the Sanskrit *chch*. I have thought it best, therefore, to represent it by these letters. This is the way in which I have dealt with the other Tamil letters, the pronunciation of which, when single, differs from their pronunciation when double;—e.g., *d*, which, when doubled, I have represented, as the pronunciation requires, as *tt*; and *ḍ*, which, when doubled, becomes in like manner *ṭṭ*.
c4.7
∗ Dr Burnell, in his "Specimens of South Indian Dialects," No. 1, Konkani (Mangalore, 1872), mentions that Professor H. H. Wilson, being accustomed to speak North Indian dialects only, used always to say *Télugu*, instead of *Telugu*.
c4.8
There is a tendency in all the Dravidian languages to pronounce *e* as if it were *ye*, and *o* as if it were *wo*. In colloquial Tamil, this pronunciation, though often heard, is seldom represented in writing; but in modern Canarese and Telugu, *y* before *e*, and *v* or *w* before *o*, are often written as well as pronounced. In Canarese and Tuḷu grammars, it has become customary, in rendering words in the Roman character, to write *ye* for *e*, and *wo* for *o*, even where the native characters employed are *e* and *o* alone—e.g., Can., *wondu*, one, and *yeradu*, two, instead of *ondu* and *eradu*. As this euphonic change seems to be a corruption, not a primitive dialectic peculiarity, and as it tends to hinder comparison with the other dialects, all such words will be written in this work without the *y* or *v*, and it will be left to the reader who is acquainted with the native usage to pronounce those words as usage requires. This usage prevails also, it seems, in Marâṭhi and Koṅkaṇi; and Dr Pope, in his "Outlines of the Grammar of the Tuda Language," points out the existence of traces of this usage even in English—e.g., "ewe" is pronounced "yew" and "one" "won." This he attributes to Celtic influences. As regards the Dravidian languages, it does not seem necessary to suppose this peculiarity to be one of any great antiquity, seeing that the spelling of Dravidian words has always been phonetic; and hence *y* and *v* would have been written as well as pronounced, if this pronunciation had been prevalent at the time the languages were first committed to writing. The people in the neighbourhood of Madura, where the purest Tamil is supposed to be spoken, pride themselves on pronouncing initial *e* and *o* pure.*
c4.9
∗ Europeans often notice the appearance of this peculiarity in the pronunciation of English by the people of South India. "Every" becomes "yevery," and "over" "wover." One of the best illustrations of this peculiarity I have heard was mentioned to me by some members of my family. As they were travelling along a road in Tinnevelly, they passed a finger-post at a cross road, on which the name of a place was inscribed in English. They did not catch the name as they passed, and therefore sent back a native girl to find it out for them. The girl knew very little English, and on her return said she could not make out the name, but could repeat the letters. "What were they?" Answer—"Yen, yeh, yell, yell, yell, woe, woe, war!" These dreadful sounds represented the name "Nalloor."
c4.10
DRAVIDIAN GRAMMAR.