[Footnote 1:  LASSEN, Indische Alterthumskunde, vol. i. p. 199, 362.]

The early inhabitants of India before their comparative civilisation under the influence of the Aryan invaders, like the aborigines of Ceylon before the arrival of their Bengal conquerors, are described as mountaineers and foresters who were “rakshas” or demon worshippers; a religion, the traces of which are to be found to the present day amongst the hill tribes in the Concan and Canara, as well as in Guzerat and Cutch.  In addition to other evidences of the community of origin of these continental tribes and the first inhabitants of Ceylon, there is a manifest identity, not alone in their popular superstitions at a very early period, but in the structure of the national dialects, which are still prevalent both in Ceylon and Southern India.  Singhalese, as it is spoken at the present day, and, still more strikingly, as it exists as a written language in the literature of the island, presents unequivocal proofs of an affinity with the group of languages still in use in the Dekkan; Tamil, Telingu, and Malayalim.  But with these its identification is dependent on analogy rather than on structure, and all existing evidence goes to show that the period at which a vernacular dialect could have been common to the two countries must have been extremely remote.[1]

[Footnote 1:  The Mahawanso (ch. xiv.) attests that at the period of Wijayo’s conquest of Ceylon, B.C. 543, the language of the natives was different from that spoken by himself and his companions, which, as they came from Bengal, was in all probability Pali.  Several centuries afterwards, A.D. 339, the dialect of the two races was still different; and some of the sacred writings were obliged to be translated from Pali into the Sihala language.—­Mahawanso, ch. xxxvii. xxxviii. p. 247.  At a still later period, A.D. 410; a learned priest from Magadha translated the Attah-Katha from Singhalese into Pali.—­Ib. p. 253.  See also DE ALWIS, Sidath-Sangara, p. 19.]

Though not based directly on either Sanskrit or Pali, Singhalese at various times has been greatly enriched from both sources, and especially from the former; and it is corroborative of the inference that the admixture was comparatively recent; and chiefly due to association with domiciliated strangers, that the further we go back in point of time the proportion of amalgamation diminishes, and the dialect is found to be purer and less alloyed.  Singhalese seems to bear towards Sanskrit and Pali a relation similar to that which the English of the present day bears to the combination of Latin, Anglo-Saxon, and Norman French, which serves to form the basis of the language.  As in our own tongue the words applicable to objects connected with rural life are Anglo-Saxon, whilst those indicative of domestic refinement belong to the French, and those pertaining to religion and science are borrowed from Latin[1]; so, in the language of Ceylon, the terms applicable to the national religion are taken from Pali, those of science and art from Sanskrit, whilst to pure Singhalese belong whatever expressions were required to denote the ordinary wants of mankind before society had attained organisation.[2]