36 A. F. R. Hoerule— ^ Collection of Hindi Roots. [No. 2, 



7. Addition of the pleonastic suffix api. Thus suMv " please," 

 Skr. suJch {as ii it were suhhdpi). In causal roots this is the universal 

 rule; e. g., Jcarav (or shortened hira) "cause to do," as if it were derived 

 from a Sanskrit root hardpi (instead of Mri). 



It will be observed that the laws 2 and 4, and again 3 and 5 are 

 closely connected. 



The preservation of a final single consonant (especially a hard con- 

 sonant) in a Hindi root is a sure sign of its having been affected by the 

 3rd or 5th law. The final g of such a very common root as lag would 

 not have been able to escape elision during its passage through Prakrit, 

 unless it had been protected by another consonant following it ; Skr. 

 lagati " he belongs" would become Pr. lacCi, H. lai ; but Skr. lagyate is 

 Pr. lagged, H. lagai or letge.*- 



The termination aya of Sanskrit roots (or rather bases) of the Xth 

 class and of causals is contracted in Prakrit to e. This e is changed to a 

 in Hindi, by the 4th law. Thus Skr. mdrgaya " ask" is Pr. magge, 

 H. manrjei (ilt^T) ; Skr. troteiya "break" is Pr. iode, H. tor a. On the same 

 principle the Skr. vilcriya "sell" (pass.), which in Pr. becomes viklce, is H. 

 hiha; thus Skr. vilcriyate "it sells," Pr. vilcleei,]^. hilcai or (contracted) S?7ce. 



Secondary roots may be divided into three sorts, according to the 

 manner of their derivpition ; whence they may be called derivative, de- 

 nominative and compound roots. 



1. Derivative roots are those which are obtained by the slwrtening 

 of a radical voioel. E. g., nali " ^o\n" from naha "bathe", Skr. snd. It 

 will be observed that this process is the exact reverse of the well-known 

 method by which Causals are formed in Sanskrit. These are made by 

 lengthening a radical vowel ; e. g., from the simple root Icar " do" Sanskrit 

 forms the causal root Icdri "cause to do," for which, by the 7th law, 

 Hindi places Jcarav or hard. Now, mistaking nalid, which really is a 

 simple root, to be a causal root (as if it meant "cause to flow"), Hindi 

 re-derives from it a simple root nah ; the pair of roots nahd and jza/t'being, 

 in outward appearance, exactly like the pair Icara and Tear. 



2. Denominative roots are made by treating nouns, as if they were 

 roots. The nouns which may be treated in this way are either substantives 

 or participles. To the former class belong such roots a,s jam " germinate," 

 derived from the Sanskrit substantive janma " birth" (of the Skr. root 

 jan "be born"). Of the other kind are ^a^YZi " enter," derived from the 



* This process is expressly mentioned by Prakrit G-rammarians, in the case of 

 a few roots; as Pr. rujjhdi (or rubbhdi) act. "he hinders" as well as pass, "he is hin- 

 dered," from Skr. pass, rucllnjate " he is hindered," while the Skr. act. is runacldld 

 (Vllth el.) ; see H. 0. 4, 218, 245, 248. But it clearly occurred in more cases, than 

 they recognized ; thus, in all those cases enumerated in H. C. 4, 230. The case of the 

 Hindi root hhaj " break" is exactly similar. See also S. Goldschmidt in J. G. 0. Soc, 

 Yol. XXIX, p. 492. and Weber SaptasataM, p. 64. 



