Page:Vedic Grammar.djvu/79

From Wikisource
Jump to navigation Jump to search
This page needs to be proofread.

II. EUPHONIC COMBINATION. CONSONANTS. 69

  • vrkans ¹; mahán for*mahánts; á-yan, 3.sing. aor. for *á-yan-st (1. sing. á-yamsam) 2.

Then became Anusvāra (or Anunäsika) before this s, which was treated in exactly the same way as when it followed an unnasalized vowel (as becoming ā, but īs, ūs, is becoming īr, ūr, Ţr). d. The an of the 3. pl. subjunctive (originally ant), however, remains unchanged before vowels within a Pāda, obviously owing to the t which at one time followed. There happen to be only five occurrences of this form under the conditions required: á vahān āšú (I. 8418); ghosan úttarā (III. 338); sphurán rjipyám (VI. 671¹); gácchăn id (VIII. 795); gacchan úttara (x. 1010) 3. 2. Before consonants. Final remains unchanged before all gutturals and labials (including m), as well as before voiced dentals (including n). It is, however, liable to be changed before the following sounds: - a. Before p, final n, when etymologically = ns, sometimes becomes mh; thus nimh pahi (vIII. 84³) 4; nrmh pátram (1. 121¹). b. Before all palatals that occur it becomes palatal ; e. g. urdhváñ caráthāya (1.36¹4) for urdhván; táñ juṣethām (v.516) for tán; vajrin chnathihi (1. 635) for vajrin śnathihis; deváñ chlókaḥ (x. 125) for deván ślókaḥ. Before c, however, the palatal sibilant is sometimes inserted in the RV., the n then becoming Anusvāra. This insertion occurs only when the sibilant is etymologically justified (that is, in the nom. sing. and acc. pl. masc.) almost exclusively (though not invariably) before ca and cid; thus anuyajámś ca (x. 518), amenámś cid (v. 31²). In the other Samhitas the inserted sibilant becomes commoner, occurring even where it is not etymo- logically justified (that is, in the 3. pl. impf., and the voc. and loc. of n- stems) 8. c. Before dental t9, final n usually remains; but the dental sibilant is sometimes inserted in the RV., the n then becoming Anusvāra. This insertion, however, occurs in the RV. only when it is etymologically justified; it is commoner in the other Samhitãs", where it appears even when not etymolog- ically justified. d. Before y r v h, final n as a rule remains unchanged; but -ān, -īn, -un sometimes become -am", -imr, -umr, as before vowels; e. g. -annam rayivý dhaḥ (VII. 91³) for -annan; dadvám va (x. 132³) for dadván; panímr hatam (1. 184²) for panin; dásyumr yónau (1. 63) for dásyun.. 12 e. Before /, final n always becomes nasalized ¹²; e. g. jigīvál lakṣám (11. 1 24). f. Before the dental sibilant, final n remains; but a transitional ¹3 may be inserted; e. g. ahant sáhasā (1. 80") 'he slew with might'; tán sám may also be written tánt sám ¹4. In the former example the t is organic; from such survivals it spread to cases where it was not justified. A similar insertion may take place before s; that is, vajrin śnathihi may become vajrin śnathihi or vajrin chnathihi (through vajriñc śnathihi for vajrint śnathihi) ¹5. ¹ Cp. Cretic ÚKOVS. 2 In all the other Samhitãs the pause forms -an -in -un -n predominate. In the post-Vedic language they became the only allowable forms. 3 Cp. OLDENBERG 428. 4 The MS. (11. 13¹¹) has ným̃ṣ pāhi. 5 For examples of nominatives with this Sandhi, see LANMAN 506A, note, 512 (top), 517. 6 Op. cit., 506 A, note, and 514. 7 Op. cit., 512. 8 There are no examples of the inserted sibilant before ch in the Samhitas. In the post-Vedic language a sibilant is invariably inserted after n before all voiceless palatals, cerebrals, and dentals. 9 Neither th nor t th occur in the Sam- hitās after final n. 10 Cp. LANMAN 516 A, note. II For dadhanvám yáḥ of the RV. (IX. 107¹) aud SV., the VS. (xIx. 2) has dadhanva yah. 12 As does before 7; see above 75, 4- 13 Before s a transitional k similarly may be inserted after a final ; e. g. pratyán sá may also be written pratyánk sá. 14 See LANMAN 506 A, note, and 346. 15 Cp. WACKERNAGEL I, 282, note.