________________
INTRODUCTION
XXXV
following that definition generally. Some times he has departed from that definition out of regard to the practice of his predecessors. Wherever he does this he gives his reasons in the Commentary. It will be seen from the above definition followed by Hemacandra in the selection of Desi' words that his object was neither philological nor historical. For this reason he had to exclude all Deśr' roots, though included by his predecessors in the list of 'Desi' words, from this collection because he had treated of them in his grammar 1 and the reason for his excluding such roots from the list of 'Desi' words and deriving them from Sanskrit roots by substitution is not that he was ignorant that they were 'Desi' but because it served the purpose of economy. With Hemacandra as with the Indian grammarians generally, brevity and the derivation of a form by rules by the shortest cut and by avoiding all roundabout processes though philologically correct and historically true, was the guiding principle. Most of the rules in Indian grammars for substitution of one form for another illustrate this principle. The substitution of 'bhū' for 'as', 'ghas' for 'ad', 'gā' for 'i', 'neda' for 'antika', 'bolla' or 'jampa' for 'katha', 'muņa' for jñā, fohira' for 'nidrā', 'kandotta' for 'utpala', 'chimchat' for 'pumscali' will show that such transformations are never meant for philological development of one form into another. That in the exclusion of 'Desi roots from the Desināmamālā and their derivation by substitution in his grammar Hemacandra was guided by the principle of economy he has made clear in the Commentary on the Desināmamālā. Similarly he includes many 'tadbhava' words in the list of Desi' words not because
1 Desīnāmamálā, 1, 3, Com.
Ibid., 1, 37. Com. Ibid., 1, 37, Com.