« AnteriorContinua »
OBs. By reason of r.30. (with note) the initial sibilant almost always takes the cerebral form sh.
103. Formation of the Base of the Third Preterite. This complex and “multiform” tense, the most troublesome and intricate in the whole Sanscrit verb, is not so much one tense, as an aggregation of several tenses, all more or less allied to each other, all bearing a manifest resemblance to the first preterite, but none of them exactly assignable to that tense, and none of them so distinct in its character or so universal in its application as to admit of segregation from the general group, under any separate title of its own.
Fortunately for the study of Sanscrit, the third preterite very rarely occurs in the earlier and better specimens of Hindu composition; and the student who contents himself with the Rāmāyana, Mahābhārata, Hitopadesha, and Laws of Manu, and avoids the grammatical Poem of Bhatti, and the extravagant writings of more modern authors, will lose nothing by an almost total ignorance of this tense, or, at least, may satisfy himself with a very cursory survey of its character and functions.
All the varieties of this tense are reducible to three distinct forms, adapted to the three schemes of terminations given in the table above. The first form is that which belongs to those roots which insert i before the terminations of the non-conjugational tenses: the second belongs to those which reject i : and the third, which most nearly approximates to the 1st pret., belongs, in the first place, to certain roots, whose bases in the first pret. present some important variation from the root; in the second, to certain roots ending in a sh, sh, or Eh, which have i, u, or și for their radical vowel; and, in the third, to causals, or verbs of the 10th conjugation.
In all the modifications of this tense, the first step in the formation of the base is the prefixing of the augment a, a further indication of its community of character with the first preterite. But besides this there are changes of the base peculiar to each form.
Form 1. Formation of the Base.
a. In the first form, if a root end in a vowel, the base must be formed in the par. by the Vriddhi change, and in the ātm. by the Guna : thus, from pū,“ to purify,” come the two bases apau and apo (wuran apāvişham, &c. r. 8., wyrafa apavishi, 8c., r. 10.). Very few roots ending in vowels take the inserted i, and therefore very few follow this form.
b. If a root end in a single consonant, Guna is required in both par. and ātm. (r.88.6.). Thus, from budh,abodh (alfri abodhisham, fc., vatfufa abodhishi, &c.). Almost all roots ending in consonants, which insert i, follow this form.
c. A medial a is sometimes lengthened; as, from vad, avād (avādisham, fc.).
d. A few roots in ā, e, o, and ai, with three in m, viz. yam, ram, nam, insert s before the terminations of this form, the final e, o, and ai, being changed to ā. Thus, from yā, ayās (ayāsisham, 8c.); from so, asās (asāsisham, &c.); from yam, ayans (ayansisham, &c.). In the ātm. they follow form 2. (ayāsi, 8c., aransi, 8c.).
Form 2. Formation of the Base.
e. The greater number of roots rejecting i follow this form.
Observe that the initial s may often be rejected from those terminations in which it is compounded with t, th. This takes place when the base ends in any
consonant except a nasal or a semi-vowel, or in any short vowel; and dhwam
(g) takes the place of dhwam when the base ends in any other vowel than ā.
f. If a root end in either a consonant or a vowel, Vriddhi takes place in the par.; as, from kri, akar (WATU akārsham, 8c.), from yuj, ayauj (wuraj ayauksham, 8c., dual ayaukshwa, ayauktam, &c.); from rudh (arautsam, &c., dual arautswa, arauddham, r. 91. a. b.); from dah (adhāksham, &c., dual adhakshwa, adāgdham, r. 91. h. ¿.).
g. But in the atm. if a root end in any other vowel than ri, Guna takes place; as, from chi, ache (wafa acheshi, 8c.); and if in ri, or any consonant, then the vowel is unchanged; as, from kri, akri (akrishi, akrithāh, 8c.); from yuj, ayuj (ayukshi, ayukthāh, 8c.); from rudh, arudh (arutsi, aruddhāh, aruddhu, r. 91. a. b.).
Form 3.* Formation of the Base. h. Roots of any conjugation, making use of this form, in general attach the terminations directly to the root: thus, from gam, agam (agamam, &c.); from bhid, abhid (white ubhidam, fc.). And no confusion can arise from this apparent identity with the 1st pret., as in all cases where these terminations are used for the 3d pret., the 1st pret. presents some difference in the form of its base; as in agachchham, abhinadam. So again, the sixth conjugation, which alone can ever shew a perfect identity of root and base, never makes use of this form for its 3d pret., unless by some special rule the base of its 1st pret. is made to differ from the root. Thus, lip, “to smear” (cf. åAelow), which is alipam in the 3d pret., is alimpam in the first.
i. Certain roots ending in long vowels, as dā, “to give,” † make use of the terminations of this form, but reject the initial vowel throughout (adām, adāh, adāt ; dual, adāva ; 3d pl. aduh ; ātm. adishi, &c., form 2.f). So bhū,“ to be," except in the 1st sing. and 3d plur. (yai abhūvam, abhūh, fc.; 3d plur. abhūvan ; in the atm. abhavishi, &c. form 2.).
j. The roots vach,“ to speak,” and pat,“ to fall,” follow this form; but, in contradistinction to the 1st pret., make their bases avoch, apapt (avocham, apaptam, cf. ÉTETTON). As, “ to throw,” makes āsth ; shās, ashish; nash, anesh.
k. Certain roots ending in sh, q sh, ph, inclosing a medial i, u, or ri, form their 3d pret. according to this third scheme; but whenever confusion is likely to arise
* Bopp has noticed that this form of the 3d pret. corresponds very clearly with the 2d aorist of the Greek (cf. asthām, asthās, asthāt with cornu, čotns, forn), and that the first two forms are more or less analogous to the 1st aorist.
† Bopp remarks that the 1st pret. of this root adadām bears the same relation to its 3d preterite adām, that édíow does to édwv. So also the relation of adhām (3d pret. of dhā) to adadhām (1st pret.) corresponds to that of conv to fríonv. Cf. also abhavah and abhūh with éques and équs.
Roots like dā, in the atm., change the radical ā to i, and follow form 2. : thus, adishi, udithāh, adita, adishwahi, 8c.
between the 1st and 3d preterites, the base is formed by the addition of sh to the root, the final of the root being at the same time changed to k (r. 91. e.).
Thus from feu dish, “ to point out,” comes the base adiksh (ufcui adiksham, fc., cf. édelga, 1st pret. adisham); from dwish, “to hate," adwiksh (adwiksham, &c.); from duh, “to milk,” adhuksh (adhuksham, &c., r. 91. i.). But this class of roots substitutes i for e, the termination of the 1st pers. sing. ātm. (adikshi, adikshathāh, &c.), and athām, ātām, for ethām, etām, in the 2d and 3d dual (adikshāthām).*
Causal verbs make use of the terminations of this form, but the base assumes a very peculiar reduplication (analogous to the Greek pluperfect), to be afterwards explained. Thus, from budh, abūbudh (abūbudham, 8c.).
yāsam yāswa yāsma || sīya sīvahi simahi
Obs. The initial sibilant, in the terminations of both these tenses, almost always takes the cerebral form șh.
Formation of the Base of the Benedictive or Precative. 104. In forming the base of this tense the parasmaipada never admits inserted i, but those roots which take inserted i in the futures (see p. 79.) take it also in the ātmanepada of this tense ; as from bhū, “to be," the two bases bhū and bhavi (apare bhūyāsam, &c., ufquia bhavishīya, fc.).
* A few roots of this kind optionally follow form 2. in the atm. ; as, fue may be अलिक्षि, अलीढाः, अलीढ, &c., and दुह, अधुक्षि, अदुग्धाः, अदुग्ध, &c., r. 91. h.i.
a. If a root end in ā, this vowel is changed to e in the par.; as, from dā, de (curi deyāsam, &c.), but not in the atm. (gratu dāsīya, &c.).
b. If in i or u, these vowels are lengthened in the par. ; as, from chi, chi ; from hu, hū (chiyāsam, &c., hūyāsam, fc.), and changed to Guna in the atm. (cheshīya, hoshīya). Similarly žand à (cari, FT).
c. If in Wri, this vowel is changed to feri in the par., but retained in the ātm.; thus, from the bases for and a (faute, quia). After a double consonant și becomes ar in the par. As also before inserted i (staryāsam, starishīya from stri). d. Ifin ?ī, this vowel is changed to īr, in both voices ; as, from
the base nit (tīryāsam, tīrşhīya); but before inserted i in the atm. to ar (tarişhīya or tarīşhīya).
e. If in a consonant, there is no change in the par., and no change in the ātm., unless the root take the inserted i, when Guna is required; as, from budh the bases budh and bodhi (budhyāsam, bodhishīya); from dwish, the base dwish (dwishyāsam, dwikshīya). But in the par. certain roots undergo changes analogous to those in the 2d pret. (r. 99. d. f.) and in the passive (r. 112. d. f.). Thus, from grah, grihyāsam, fc.; from vach, uchyāsam, &c.; from swap, supyāsam ; from yaj, ijyāsam, &c.
It is to be remarked of this tense, that the changes of the base before the y of the par. terminations are analogous to those before the y which is the sign of neuter and passive verbs (r. 89. 112.). Observe, moreover, that the terminations resemble those of the potential of irregular primitives (p. 66.); the only difference being, that in most cases a sibilant, and in some, two sibilants, are either prefixed or inserted.
Formation of the Base of the Conditional. 105. This tense, in its formation, lies half way between the first preterite and the second future. It resembles the 1st pret. in prefixing the augment a to the base, and in the latter part of its terminations : it resembles the second future in inserting i in exactly those roots in which the future inserts i, and in the first part of its terminations. Thus, from budh comes the base abodhi (valfya abodhishyam, fc.); from kship, the base akshep (wari akshepsyam, fc.).
Formation of the base. 106. The base of the infinitive is identical with the base of the first future, and where one inserts i, the other does also. Thus from budh, bodhi (alfug bodhitum); from kship, kshep (an ksheptum). Moreover, all the rules for the change of the final consonant of a root before the t of the future terminations apply equally before