various senses in which this tense is now employed, as the point is one which belongs not to the domain of comparative philology, but to the grammar of each individual language. The name "aorist," which I have suggested, has the advantage of being indefinite as to time, and in this way represents fairly scope of the tense.

the

§ 34. The next simple tense is the IMPERATIVE, and this, like the aorist, is descended from the imperative of the ancient languages. As might be expected, it closely resembles the aorist or old present, and has the following scheme of endings:

In this scheme only the second persons singular and plural have been given for P. S. and G., because the other persons are the same as the aorist. This is also true of H., the aorist being used as a potential in all these languages, the first and third persons of both numbers can only be considered imperatives in so far as the potential is itself imperative, just as in Latin and other Indo-European languages. So we may say in H. parhe "he reads," or, "let him read." It was shown in §§ 4, 5, etc., that even in Pali and the Prakrits the present and imperative had been confounded together, a practice that has paved the for the modern system.

way

It is only in M. O. and B. that the third singular has a

This mark means that the 2 sing. is the simple stem, as kar" do thou!" parh "read thou!"

separate form, which may in all cases be traced back to the Skr. 3 sg., Pr. 3, which in M. becomes. In M., however, the termination for this tense is also in use. To the same origin may be ascribed the O. and B., the final of which, however, presents considerable difficulty. It will be discussed along with a similar termination in the future.

The third plural in M. O. and B. is parallel to the singular, and is connected with Skr., just as the corresponding person in the aorist is with Skr. f. In O., owing to the influence of the final u, this termination is often written untu, as karuntu "let them do," jâüntu "let them go."

In all but S. the second singular is the bare verbal stem. In M. a final is heard, and slightly also in B. and O. In the dialect of Northern Gujarat a is sounded after the final consonant, as "do thou," "speak thou," "go thou.”1 But in the rest this person ends with the final consonant, as kar "do," dekh "see." In the H. mediæval poets this person often ends in fa, as stated in §§ 4, 5, 7, corresponding to which is a plural in, as

तुम्ह लेहु तुम्ह लेह मुष जंपि जोध ॥

"Seize ye! seize ye! muttering of war."-Chand, Pr. R. xix. 33.

This form is also found in G., and in Old-M. takes the shape of एं, as पावें “ find thou !” for पावई, from पावहि, with inorganic anuswâra. Sindhi, which causes all its words to end in a vowel, makes this person end in, which is apparently only a weakening of the final vowel of the stem. The dialectic forms of H. present few noteworthy peculiarities, in some cases the forms which Kellogg gives as imperatives are really other tenses used imperatively. Thus the form “beat ye!” common in the eastern area, is really a future, "ye shall beat.” Often, too, in colloquial Hindi, and in Urdu, in giving an order,

1 Vans Taylor, Grammar, p. 89.

the future is used, as राम को अपने संग लाओगे “You will bring Râm with you," that is, "bring him with you!" So also the infinitive, as यह सब काम आजही करना “ Do all this work today,” literally, “(Take care) to do," cu fiu âî ya â aa "Pay this debt, and then go away."

Most of the seven languages have, in addition to the ordinary imperative, a respectful form used in addressing a superior, or in entreating and asking a favour. This, in Hindi, ends in Sing., Plur. t. In P. this form is seldom employed, and when used, may be considered as borrowed from H. In the other languages are

In a few stems in H. which end in e, is inserted between the stem and the termination, the final vowel being changed to ¿, as ले “take,” लीजिये, दे “give," दीजिये ; the stem कर “do” is in this case changed to, making af "be pleased to do." Sindhi sometimes takes in the singular instead of a, probably on the analogy of the simple imperative, which ends in ¬; and in the plural, instead of इजो, the forms दूजा, इजार, इजाह, इजाहु are used when great respect is implied, as वञिज़ा "be pleased to go," "be pleased to hear." Many of the rustic dialects of Hindi have also this form; thus Rajputana दूजे, इज्यै, ईजे, or simply जे, as मारिज्ये, मारीजे, मार्जे “be pleased

to strike."

Vararuchi (vii. 20, 21, 22) teaches that jja and jÿâ may be optionally substituted for the affixes of the present and future, also for those of the imperative, in verbs which end in a vowel. In Old-Marathi, accordingly, a form with inserted is found in present, past, and future, as well as imperative, as aftan “ he does,” करिजेला “he did,” करिजेल “ he will do, ” करिजे "do

1 Pincott's Sakuntalâ, p. 12, a first-rate text-book in admirable idiomatic Hindi.

and

thou," in which the junction vowel between the inserted the termination has been changed to e. As, however, the inserted is also a type of the passive, this form has occasionally been mistakenly used in a passive sense, as frant “I am

struck." Lassen (p. 357) refers this increment to the Skr. potential, which is confirmed by the Pali forms quoted in § 4, and by the dotted in S., which usually indicates that a double letter has existed. The of the Skr., as in qati, is doubled in Pa. f, and hardened to in Pr., whence the modern, with lengthening of the preceding vowel in H., and change of e to a in G. (cf. G. ¿e=za). As Vararuchi, in extending the use of this increment to present and future, is writing of the Maharashtri dialect, it is not surprising that the modern Marathi should show a wider use thereof than the sister idioms, in which the sense of a potential has passed over into that of a respectful imperative, or, as Trumpp well calls it, precative.

To this tense belong the two M. words म्हणजे and पाहिजे. The former is the precative of ¶ (Skr. √ н) "to speak," and means literally, "be pleased to say;" in modern times it means "that is to say," "i.e.," "videlicet," as A FEUÂ DIEI अश्व म्हणजे घोडा "açwa, that is to say, a horse." It has also a future form म्हणजेल, meaning “in that case,” as पाऊस पडला म्हणजेल "If rain falls, then (or, in that case) there will be a crop.” The latter, with a plural पाहिजेत, and a future पाहिजेल, is from "to see," literally "please to see," and means "it ought," as a ad ufga “this work ought to be done," literally "please to see (that) this work is done;""see being used in M., as in English, in the sense of "seeing to," "providing for," "taking care for."1

Similar to these is the H., lit. "please to wish,” but

1 See Molesworth's Marathi Dictionary under these words; also Godbol's Marathi Grammar, p. 92.

meaning "ought," and, like fa in M., used with the past participle, as यह काम किया चाहिये “ this work ought to be done." Colloquially, however, and even among good writers,

fa is often, like other ancillary verbs, constructed with the oblique form of the infinitive, and it would not be absolutely incorrect to say. In fact, the construction both in M. and H. with the past participle remounts to a period when the participial character of this form was not yet forgotten. Since, however, the past participle in H. has come to be used simply as a preterite, this construction has lost its significance. Not so in M., where, as will be seen hereafter, the distinction between the preterite and participle still survives.

Gujarati has an analogous formation in the word

“it

is wanted," French "il faut," Italian "bisogna." It is from the verb "to see," and is used with a dative of the subject, as मनें बीजो वर जोये नहीं “I want no other blessing,” like Latin "mihi necesse est, oportet, decet," etc. It is conjugated throughout the full range of tenses, as C Hİ Â HİC जोइतुं हतुं “Whatever was required for the voyage,” जो मारे हरण जोशे तो " Should I require venison, then

.

§ 35. A simple future derived from the old synthetical tense exists only in Gujarati and in Old-Hindi.

follows, taking the stem kar "do," as a type:

The tense is as

Old-H. after afte aft?

aft aftet afte.

Kellogg (Gr. p. 238) gives the following interesting transitional type from eastern Rajputana :