The Ādi Granth, or the Holy Scriptures of the Sikhs (1877) · 42

The Present tense offers a great variety of personal terminations, many of which coincide, the Anusvara being generally dropped in the writing of the Granth, so that the person can in many instances only be found out by the context. The Future offers two forms, one of which is made up by the terminations?TT. TfvJ>?T\, etc., corresponding to the Prakrit terminations T^FTrf'T, Wflf. Wf?. etc., and the other of which is a compound tense, as now used in Hindi and Hindustani (see my Sindhi Grammar, p. 291, annotation).

The Preterite is made up either by means of the participle past alone or by personal termina- tions added to it (as in Sindhi).

There is still a regular passive voice for the Present, Imperative, Future, and paiily also for the Preterite, which has been lost already (with a few exceptions) in the later Hindu!, which shows the first attempts to make up for it by having recourse to a composition.

There is not only a participle present active, formed by the terminations " anto," "ant," " atu," "at," "td" or "da," but also a participle present passive, formed by the termination "latu," "lat." The participle past offers still the older form "id" [id iau); many are also directly taken over from the Prakrit without any further assimilation. To this must be added, that pronominal suffixes are not unfrequently connected with the verb, especially in the Preterite, more rarely so with the noun. In Sindh! this, is still very common and quite a peculiar feature of this idiom, but in the later Hindu! only a few traces of it are to be mot with, and in Hindi and Hindiistaal it is

From these few remarks it may be inferred, that the idiom of the Granth is well worth a closer investigation, as we shall thereby get a clearer insight into the formation of the modern Korth-Intlian vernaculars, the peculiarities of which cannot be fully laid open without going back to the source from which they have sprung, and which has fortunately been preserved to us in the Granth.

The whole Granth is written in verses, as the Hindus have very little taste for prose-compositions. The artificial measures of Sanskrit poetry are all discarded; the metres that are used in the Granth, are either old Prakrit metres or later inventions, perhaps of the poets themselves. There are two leading principles in Hindu! poetry, viz.: the verses are measured by quantity only, i.e. by the

This mood (though originally the Preseut tense) is in our modern Hindi and Hindustani grammars generally called the Aorist, which is quite an inappropriate appellation. In native Hindustani grammars it is designated I)y c iLi<, as it in some way corresponds to the Arabic c jLi-*, which De Sacy translated by Aorist. Thence it was applied also to the Hindi and Hindustani. But it was quite lost sight of tliat the technical terms of Arabic grammar cannot be properly applied to the Hindustani verl), which in its conjuga- tion is totally different from the Arabic verb; and if tlie Maulavis term this tense or rather mood (as it is now) c,Li», the Arabic not offering a more suitable appellation, we at any rate must not render it by "Aorisi," as this term conveys a different meaning altogether.

cxxviii ON THE LANGUAGE AND THE METRES USED IX THE GRANTH.

the number of moras (aot by number of syllables or fixed feet), and tbey must rhijme together; the metres are therefore all so-called mdtra chandas (regulated by quantity), as they are intended for singing' or a rhythmical recitation. The greatest attention is paid to the rhyme, as in our modern poetry; and if- the poet cannot command it readily, the last word is tortured into it and thereby frequently so disfigured, that its original form is hardly recognisable.

It cannot be our object here to give a full description of all the metres employed in the Granth, as this would cany us beyond the limits of a preliminary discourse; we must content ourselves with laying down here the general rules and giving a survey of the most common metres, which will enable the student to find out by scanning those fancy metres of some poets, to which no name has been given in the Granth, as very likely no name was known for them, whereas the usual metres are always indicated by their appropriate names. The Sikhs themselves seem now to have lost all knowledge of the metrical laws of the Granth, for I never met a person who could give me the least clue to them, and the learned Brahmaus disdain to read the Granth.

The length of the syllables is determined by allotting to a short syllable one matra or kala, i.e. mora, and to a long syllable two matras. A syllable is long either by nature (as d, f, ft, e, o, ai, an) or by position (when a naturally short vowel is followed either by a double or a conjunct consonant); but if the second part of a conjunct consonant be a semi-vowel, a k, jJ, i or h, the pre- ceding vowel may remain short, if the metre requires it. On the other hand, a single consonant may be doubled, in order to gain a long syllable; this is especially the case, if it be originally doubled in Prakrit; for instance f^^T tisu may also be pronounced tissu (Prakrit fTTOT). To this cir- cumstance especial attention must be paid, as the doubling of a consonant is never indicated in the writing of the Granth, not even in such cases where in common pronunciation it is still doubled.

A diphthong may again be severed into two short syllables, as ai = a-i, au = a-u, wherever required by the metre. A naturally long vowel may be pronounced sJioft; this is especially the case with e and o, which may be considered as anceps, as in Prakrit; even the diphthongs ai and au may be rendered short under the pressure of the metre. On the other hand, the poets take the liberty of lengthening a short vowel, whenever necessary for the sake of the metre, not only at the end of a verse or hemistich, but also in the midst of a verse, though this is comparatively seldom the case. Two vowels may be contracted into one, especially i with a following vowel, i being changed in this case into the semi- vowel y, as itc = i/u; even a short and a long vowel may be contracted, as ai to ai (short or long, as the metre may require it).

Another point, which must well be attended to, is, that the pronunciation of the Hindu! difi'ers greatly in poetry from that usual in prose. In prose the consonants are now frequently mute, and so always a final consonant containing short a; but in scanning a verse, no vowel is, as a rule, to be passed over; even a conjunct consonant must now and then be separated into its constituent parts (by the insertion of i or a), in order to gain the necessary number of matras.

The Dohra, which is comparatively little used in the Granth, is a distich, the two verses of which rhyme at the end. Each verse consists of twenty-four matras, which are distributed into feet (3n!r) of 6 -1- 4 -}- 3 and 6 -f 4 -)- 1 matras respectively, there being a caesura at the end of every

' III the Granth the key-note (UfJ) is therefore generally added, as the verses of the Granth are still sung in the public worship of the Sikhs, especially in tlie Har-niandar at Amritsar, accompauied by stringed musical instruments.

' Sanskrit f^T?^, Prakrit ^fT; dohara (usually now pronounced dohra) is tlie dhninutive of it. The Doha is an old Prakrit metre, as it is already found in Kalidasa's VikramorvasI (edited by Bolleuscn, St. Petersburg, 184G), p. 55 and p. 3/3.

ON THE LANGUAGE AND THE METRES USED IX THE GEANTH. exxix

first hemistich. The syllables of a whole Dohra may therefore be from forty-eight short ones to twenty-three long and two short ones, the final matra of each verse being always short. We subjoin here two Dohras of Teg-bahadur (Transl. p. 708), in order to show their scansion.

bala chutkio | bandana | pare || kachu naho|ta upa|i||

kahu nanaka | aba o|ta hari || guja jiu ho|hu saha|i || v. 53.

^f3T?TMT Tf^ 3ffT 31^ 7# 7J fXyfoG?rr^ ii

sangi sakha | sabhi taji ] gae || kou na nib|hyo sa|tha ||

kahu nanaka | iha bipa|ti mai || teka eka | raghu najtha || v. 54.

The Sorattha is the reverse of the Ddhra, the smaller half preceding the longer one; it is very seldom met with in the Granth.

The Dupada is of very frequent occurrence in the Granth; it consists usually of a distich of two verses, which rhyme at the end, each verse containing thirty-four matras, and being divided by a caesura into two equal hemistichs. The verse is distributed into feet of2xG-|-4 + 5-(-2 Kalas, the last syllable always being long. We subjoin an example taken from Eag Gaurl, Mah. V., Sabd 115, 2 (Transl. p. 291):

dina na|tha ana|tha karuna | mai || sajana milta pita | mabitarl|a || 17 -j- 17 =: 34 K.

eUTJ 7^^ ftrg^ 3lfTJ TJTASn t VWS ^3 VTfH €3ftnn II

carana kavala | hirdai | gahi nanajka || bhai sagara | eanta pa|ri utarl|a || 17 -}- 17 = 34 K.

The Dupada presents many varieties. One kind consists of only one verse, the two hemistichs of which rhyme together, and are also written as separate verses and counted as such. In this case the verse is scanned by6-l-4 + 5-|-2|[6-t-4-f4-f-2=33 Kalas, orby6 + 4 + 3-f2[|6+4 + 4 + 2 = 31 Kalas, the last syllable always being long. Now and then we meet with Dupadas, which alternately contain 34, 33 and 31 Kalas. Usually a Rahau also is added, scanned in the same way, but counted as a separate verse. We subjoin here an example, taken from Rag GaurT, ilah. V., Sabd 114 (Transl. p. 291):

^nr 7 v^ -^{37; ^rs^ ii 5T3 f^;^ t; fjwr^ sjtc^ ii ■3U^^Q ii