I. PHONOLOGY. THE CEREBRAL. ANUSVĀRA AND ANUNĀSIKA. 53 3. The various origins of h led to some confusion in the groups of forms belonging to roots which contain h. a. roots in which represents an original guttural (IIr. jh) show some forms which would presuppose an old palatal (IIr. 2). Thus from muh- 'be confused', phonetic derivatives of which are mugdhá-, mógha-, are formed mūḍhá- (AV.) 'bewildered' and múhur 'suddenly'. b. roots in which represents an old palatal (IIr. h) show forms with a guttural before vowels and dentals; thus from dih- 'besmear' is formed digdhá- (AV.) 'owing to the influence of dah- 'bum', and duh- 'milk'. Similarly, froin *saráh- 'bee', beside sarádbhyas, is formed säraghá- (AV. VS.) 'derived from the bee'¹. c. the root ruh-, though-rudh- 'ascend', is treated as if the represented (as in vah-) an old palatal (IIr. zh); hence aor. áruksat, des. rúruksati, Pp. -rūḍha- (AV.), ger. rūḍhvá (AV.). 59. The cerebral .This sound, as distinguished from the ordinary dental, is a cerebral ². It appears in our (Sakala) recension of the RV. as a substitute, between vowels, for the cerebral d and, with the appended sign for h, for the cerebral aspirate dh³; e. g. ila- 'refreshment'; á-şalha- 'invincible'. It also appears in Sandhi when final before an initial vowel; e. g. turāṣál abhibhutyojaḥ (III. 434). In the written Samhitā text, however, it does not appear if followed by a semivowel which must be pronounced as a vowel; e. g. vidv-ànga- 'firm-limbed', to be pronounced as vilu-anga- and analysed thus by the Padapatha. In one passage of the RV. the readings fluctuate between ! and / in nilavat or nilavat (VII. 976)4. 60. Anusvāra and Anunāsika. - Anusvara, 'after-sound', is a pure nasal sound which differs from the five class nasals; for it appears after vowels only, and its proper use is not before mutes, but before sibilants or h (which have no class nasal). But it resembled the class nasals in being pronounced, according to the Prātiśākhyas of the RV. and the VS.5, after the vowel. The vowel itself might, however, be nasalized, forming a single combined sound. The Pratiśakhya of the AV.6, recognizes this nasal vowel, called Anunāsika, 'accompanied by by a nasal', alone, ignoring Anusvāra, which with the preceding vowel represents two distinct successive sounds. WHITNEY', however, denies the existence of any distinction between Anunāsika and Anusvāra. In any case, the Vedic Mss. have only the one sign ³ (placed either above or after the vowel) for both³, employing the simple dot - where neither Anusvāra nor Anunāsika is allowable. The latter sign is used in the TS. for final m before y v, and in both the TS. and MS. before mutes instead of the class nasal9, a practice arising from carelessness or the desire to save trouble. There is thus no ground for the prevalent opinion that ³ represents Anunāsika and Anusvāra ¹0. Throughout the present work m with a dot (m) will be used for both, except if they appear before a vowel, when m with the older sign (m) will be employed. 10 29. I The root sagh- 'take upon oneself',| 4 Cp. BENFEY, Gött. Abhandl. 19, 138, has no connexion with sah 'overcome'; | note. See also WHITNEY on APr. nor väghát- institutor of a sacrifice', with vah 'convey'. The relation of gó-nyoghas- 'streaming among milk', to vah- is un- certain; cp. WACKERNAGEL I, 220 b, 5 See RPr. XII. 13; VS. I. 74f., 147 f. 6 See WHITNEY on APr. I. II. note. 2 Cp. above 42. 3 This practice is followed by the Mss. of secondary texts of the RV. 7 On TPr. II. 30; JAOS. 10, LXXXVIf. 8 Cp. WHITNEY on TPr. II. 30. 9 In the MS. also when -am is written for -ān. 10 See WHITNEY, JAOS. 7, 92, note.
