Page:Vedic Grammar.djvu/326

 316 I. ALLGEMEINES UND SPRACHE. 4. VEDIC GRAMMAR. the root duh- 'milk' is dóh-a-; of yuj- 'join' yunáj-a-; but of bhu- 'be' bháv-ā-. Owing to the analogy of the a- conjugation, other verbs sometimes add a instead of a, e. g. brav-a-thá frcm bru- 'speak". The subjunctive is on the whole inflected like an indicative, but with fluctuations between the primary and the secondary endings, besides some variations in the endings themselves. Thus in the active, (r) the ending of the 1. sing. is -āni, of which the ni is dropped thirteen times in the RV., e. g. dóh-āni, yunáj-āni, bháv-āni; bháv-ā; (2) the 1. du. and 1. 3. pl. have the secondary endings -va, -ma, -an only; e. g. dóh-āva, dóh-āma, dóh-an; bháv-āva, bháv-āma, bháv-ãn; (3) the 2. 3. sing. may take the secondary endings as well as the primary; e. g. dóh-a-si or dóh-a-s; bháv-ā-ti or bháv-ā-t². In the middle, (1) the only secondary ending is found in the 3. pl., -anta, which occurs beside and more frequently than -ante; (2) the ending -ai, which is normal in the 1. sing. (being = a +e), has spread from that person to forms in which e would be normal. Thus the 1. du. has -āvahai only; in the 1. pl., -āmahai is the usual form in the RV. and AV. beside the rarer -amahe; in the 2. sing., -sai always appears for -se in the AV., though it does not occur in the RV.; in the 2. pl., -dhvai occurs once for -dhve³ in the RV:; in the 2. 3. du., -aithe and -aite occur several times in the RV., being doubtless intended for subjunctive modifications of the indicative -ethe and ete of the a- conjugation; in the 3. sing., -tai occurs once in the RV. for -te, and is the usual form in the AV. sing. a. The subjunctive endings in combination with the -a of the stem are accordingly the following: Active du. -ā-va pl. sing. -ā I. āni, -ā-ma I. -ai 2. -a-sis, -a-s6 -a-thas -a-tha 2. -a-se, -a-sai 3. -a-ti, -a-t -a-tas -a-n 3- -a-te, -a-tai 415. Injunctive. The unaugmented forms of past tenses used modally, are sometimes called improper subjunctives, but they are more suitably termed injunctives, as they appear to have originally expressed an injunction. This is borne out by the fact that since the IE. period the second and third persons imperfect (except the 2. sing. act.) had come to be used as regular imperatives expressing a commande. But the unaugmented forms of the imperfect that could be distinguished from the regular imperative (as bháras, bhárat, bharan) and especially unaugmented aorists", are often used in a sense fluctuating between that of the subjunctive (requisition) and of the optative (wish)". Thus bháratu 'let him bear', but bhárat 'may he bear', bhút 'may he be'. Middle du. pl. -ä-vahai -a-mahai, -ā-mahe -aithe -a-dhve, -a-dhvai -a-nte, -a-nta. -aite 416. Optative. This mood, which is comparatively rare in the Sam- hitas, is formed from the present, the perfect, and the aorist. The stem is formed with -ya or -ī, which, when strong and weak stem are distinguished, ¹ Cp. WHITNEY 560e; BRUGMANN, KG.719. 2 The subjunctive in à is in origin an old injunctive: BRUGMANN, KG. 716 (end). 3 In the form mādayādhvai 'may ye rejoice'. 4 kynvaite, however, appears once as an indicative; see DELBRÜCK, Verbum p. 45. 5 In the aor. subj., -si occurs only once in the RV. 6 In the a-conjugation à appears throughout: -āsi, -ās, etc. 7 In the aor. subj., -ti occurs only six times in the RV. 8 WHITNEY 563. 9 The 3. sing. and pl., e. g. bhárat-u and bhárant-u, are explained as injunctives and the particle : BRUGMANN, KG. 729, 1. 10 The aorist injunctives were probably used originally with the prohibitive particle má only: BRUGMANN, KG. 716, 2 (end). ¹¹ Cp. WHITNEY 575.