The First and Second Future of Causals.

In these tenses the inserted i is invariably assumed between the base, as formed in the conjugational tenses, and the usual terminations. Thus, from budh, bodhayi (bodhayitāsmi, &c.; bodhayiṣhyāmi, &c.).

The Third Preterite of Causals.

109. In the formation of the base of this tense, the affix ay is rejected; but any other change that may take place in the conjugational tenses, such as the insertion of p or y, is preserved. The base is a reduplicated form of this change, and to this reduplication the augment a is prefixed. Thus, taking the bases bodhay and jāpay (caus. bases of budh, "to know," and ji, “to conquer ”"), and rejecting ay, we have bodh and jāp; and from these are formed the bases of the 3d pret., abūbudh and ajijap (¶¶¶ abūbudham, &c., ai ajijapam, &c., cf. the Greek pluperfect). The rule for this reduplication is as follows:-The initial consonant of the root, with its vowel, are reduplicated, and the reduplicated consonant follows the rules given at r. 99.; but the reduplication of the vowel is peculiar.

Rules for the Reduplication of the Vowel of the Initial Consonant.

Causal bases, after rejecting ay, will end either in āy, āv, ār, or a consonant preceded by a, a, e, o, or ar. The usual reduplicated vowel for all these vowels except o, is i. But u is reduplicated for o, and sometimes also for āv. In general, this reduplicated vowel is made long, and to compensate for this, the long vowel of the causal base shortened, or, if it be Guna, changed to its cognate short vowel. Thus, the causal base nāy (from ni, rejecting ay) makes the base of the 3d pret. aninay (anīnayam, &c.); the causal base bhāv (from bhū) makes abībhav; the causal base kār (from kri), achikar; gam (from gam), ajīgam; pach (from pach), apīpach; pāl (from pā), apīpal; ved (from vid), avīvid; vart (from vṛit), avivrit. But bodh (from budh), abubudh; and sāv (from su), asuṣhav. Sometimes the reduplicated vowel is only long by position before two consonants, the radical vowel being still made short; as, shrāv (from shru) makes ashushrav; drāv (from dru), adudrav; bhrāj, abibhraj. Sometimes the reduplicated vowel remains short, whilst the vowel of the causal base, which must be long either by nature or position, remains unchanged. Thus, the causal base jiv (from jiv) may make ajijiv; chint, achichint; kalp, achikalp. In such cases a may be reduplicated for a or ā; as, laksh makes alalaksh; yāch, ayayāch; vart (from vṛit) avavart, &c.†

↑ The following bases of the 3d preterite are formed anomalously from causal

Rules for the Reduplication of an Initial Vowel.

If a root begin with a vowel this vowel is not reduplicated, only coalescing with the augment a according to the rule, p. 64.; but the reduplicated form of the final consonant, with the vowel i, is inserted between the augment and vowel thus coalescing, and the final consonant. Thus the root aksh makes achiksh p. 75. b. (āchiksham, &c.); arh, ārjih; āp, āpip; īḍ(§3), aiḍiḍ; ûh, aujih; ṛidh, ārdidh.

The Benedictive and Conditional of Causals.

110. The base of these tenses does not differ from that of the non-conjugational tenses; ay is retained, and after it the inserted i invariably assumed; excepting in the benedictive paras., where both ay and i are rejected. Thus, from budh, the bases bodh, bodhayi, abodhayi (bodhyāsam, &c.; bodhayishiya, &c.; abodhayishyam, &c.).

PASSIVE VERBS.

Every root in the language may take a passive form. It is a form, however, little used, except in the 3d person sing. and plural of the present and imperative; for although a passive construction is exceedingly common in Sanscrit syntax, yet almost all the tenses of the passive verb are expressed by participles.

Our reasons for denominating the passive a distinct derivative from the root rather than a voice of the primitive, and for considering that the 4th conjugation of primitives results from the occasional assumption of a parasmaipada form* by this passive verb, have been already givent (see pp. 58. 62.).

111. Passive verbs are conjugated with the regular ātmanepada

bases, apiyy from pay (pā, “to drink "); atishthip from sthāp (sthā, “to stand"); adhyajigap from adhyāp (i, "to go," with adhi); ajighrip from ghrāp.

*The 4th conjugation can hardly be said to

possess an ātmanepada; or if so, And it seems probable that those

then its ātmanepada is identical with the passive. ātmanepada verbs, such as pad, "to go,” and budh, "to know," which are placed under the 4th conjugation, are in reality passive verbs; at any rate, the forms given for their 3d preterites (apādi, abodhi) can only belong to passives.

That the passive does occasionally take the terminations of the parasmaipada is corroborated by Bopp, who gives several instances; as, chhidyet for chhidyeta. Nal. xiv. 6.; mokshyasi for mokshyase, “thou shalt be liberated." Other instances may be found in Westergaard; as, vidyati for vidyate.

N

inflections; that is, in the conjugational tenses they conform to the scheme for the atm. at p. 63., and in the non-conjugational tenses to the general schemes at p. 73., p. 81., and p. 84.

In the third preterite they take the first two forms (p. 81.), according as the root may admit the inserted i or not; but require that, in the 3d pers. sing. of both forms, the termination be invariably i (₹) in place of iṣhṭa and sta.

Conjugational Tenses.

112. In the formation of these the passive verb is to be treated like a regular primitive of the 4th conjugation; that is, in general the only change made in the root is the affixing of y; but before this affix certain changes of a final vowel may take place, some of which are analogous to those of the 4th conj. A final ā, e, ai, or o are changed to : as, from da, the base day (3d sing. a); so also dhā, pā, sthā, hā, mā; so also gai, "to sing," (gły). a. If in i or u, or a semi-vowel preceded by i or u, these vowels are lengthened: as, from ji, jiy; from hu, hūy; from div, divy.

b. If in ri, this vowel becomes ftri; as, from, the base kriy (f); but ar, if two consonants precede; as from smri, smary. c. If inr, this vowel becomes īr: as, from kri, "to scatter," kiry, &c.). But from pri, pūry.

कॄ

d. Roots ending in two consonants, of which the first is a nasal, usually reject the nasal: as, stabhy; from sanj (), sajy. e. Jan, khan, and tan, may optionally reject the final nasal, but the a is then lengthened: as, from jan, jāy or jany (♬ or ).

from bandh, badhy; from stambh, From comes f(p. 70. e.). शास्

जायते जन्यते

f. From vach, vad, vap, vas, swap, come the bases uchy, udy, upy, ushy, supy: from grah, prachh, and vyadh, the bases grihy (HÀ, &c.), prichhy, vidhy: from yaj, hwe, ve, come ijy, hūy, ūy.

Non-conjugational Tenses.-Second Preterite of Passives.

The base of this tense in the passive verb is identical with that of all primitive verbs. The bases, therefore, as formed at p. 74., will serve equally well for the 2d preterite of the passive, provided only that they be restricted to the atmanepada inflection.

The First and Second Future of Passives.

113. In these and the remaining tenses no variation can occur

from the bases of the same tenses in the primitive, unless the root end in a vowel. In that case the insertion of i may take place in the passive, although prohibited in the primitive, provided the final vowel of the root be first changed to its Vriddhi substitute. Thus, from chi, to gather," may come the base of the 1st and 2d fut. pass. chāyi (chāyitāhe, &c., chāyiṣhye, &c.), although the base of the same tenses in the primitive is che (chetāhe, &c., cheshye, &c.). Similarly from hu and kri may come hāvi and kāri (hāvitāhe, kāritāhe), although the bases in the primitive are ho and kar. In like manner i may be inserted when the root ends in long ā, provided that, instead of a change to Vriddhi (which is impossible), y be interposed between the final a and inserted i. ā Thus, from da, "to give," may come the base of the fut. pass. dāyi (dāyitāhe, &c.), although the base of the same tenses in the primitive is dā (dātāhe, &c.). But in all these cases it is permitted to take the base of the primitive for that of the passive, and chetāhe or chāyitāhe may equally stand for the 1st fut. pass.

*

In the case of roots ending in consonants, the base of the two futures in the passive will be identical with that of the same tenses in the primitive, the inflection being that of the ātm.

The Third Preterite of Passives.

114. In this tense, also, variation from the primitive may occur when the root ends in a vowel. For in that case the insertion of i may take place, although forbidden in the primitive, provided the final of the root be changed to Vriddhi. Thus, from chi may come the base of the 3d pret. pass. achāyi (achāyishi, &c., form 1. p. 81.) although the base in the ātm. of the primitive is ache (acheshi, &c., form 2.). So also, from hu and kri may come ahāvi and akāri (ahāviṣhi, akāriṣhi, form 1.), although the bases in the atm. of the primitive are aho and akri (ahoshi, akṛishi, form 2.). Again, i may be inserted when the root ends in long ā, provided that y be interposed between final ā and inserted i. Thus, from dã may come adāyi (adayishi, &c.), although the base in the atm. of the primitive is adi (adishi, &c.).

* This explanation of the passive, although at variance with that of Wilkins and Bopp, rests on the authority of Pāṇini (6. 4. 62.), and the Siddhānta kaumudī.

† The root दृश्, however, in the passive, may be दर्शिता हे, दर्शिष्पे, as well as द्रष्टाहे, द्रष्ट्ये ; and हन् may be घानिताहे, घानिष्ये, as well as हन्ताहे, हनिष्ये; and मह् may be माहिताहे, माहिये, as well as महीताहे, महीष्ये.

But in all these cases it is permitted to take the base of the primitive for that of the passive (so that the passive of chi may be either achāyiṣhi or acheshi), except in the 3d pers. sing. where the terminations ishta and sta being rejected, the base, as formed by Vriddhi and the inserted i, must always stand alone; thus, achāyi, “it was gathered"; ahāvi, "it was sacrificed"; akāri, “it was done"; adāyi, “it was given.” If the root end in a consonant, the base of the 3d pret. pass. will always be identical with that of the 3d pret. atm. of the primitive, except in the 3d pers. sing., where i being substituted for the terminations ishta of the 1st form and sta of the 2d form, invariably requires before it the lengthening of a medial a, and the Guna of any other medial vowel.* Hence, from tan, the form of the 1st, 2d, and 3d sing. 3d pret. will be atanishi, ataniṣhṭhāh, atāni; from kship, akshipsi, akshipthāh, akshepi; from vid, avedișhi, avediṣhṭhāh, avedi. This 3d pers. sing. of the 3d pret. passive is not unfrequently found, even in the simplest writings.

The Benedictive and Conditional of Passives.

In these tenses the same variation is permitted in the case of roots ending in vowels as in the last; that is, the insertion of i is allowed, provided that, before it, Vriddhi take place in a final vowel capable of such a change, and y be interposed after final a. Thus, from chi may come the bases chāyi and achāyi (chāyiṣhīya, achāyiṣhye); from hu, hāvi and ahāvi; from kri, kāri and akāri; from dā, dāyi and adayi. But cheṣhiya, acheṣhye, hoṣhiya, ahoṣhye, &c., the forms belonging to the atm. of the primitive, are equally admissible in the passive.

Passive Infinitive Mood.

There is no passive infinitive mood in Sanscrit distinct in form from the active. But although the affix tum has generally an active, it is capable of a passive sense, when joined with certain verbs, especially with shak, "to be able." In the Hitopadesha it is also used passively, in connection with the participles ārabdha and nirūpita.

Passive Verbs from Causal Bases.

115. In forming a passive verb from a causal base, the affix ay is rejected, but the other changes of the root are retained. Thus, from the causal base pātay (root pat, “to fall”) is formed the passive causal base pāty (pātye, “ I am caused to fall"); so also, from sthāpay (sthā, “to stand"), sthāpy (sthāpye, “I am made to stand"); from pathay (path, "to read"), pāṭhy (pāṭhye, "I am taught to read") from ("to know"), ("I am caused to know.")

* This rests on the authority of Panini, the Siddhānta kaumudī, and the Bhatti kavya (15. 64, 65.).