8
INTRODUCTION.
The number of verses is slightly different in A, B and C. Each MS. omits some which occur in the other two (A 27, B 17, and C 34), and B has eight which do not occur either in A or in C. This test, though indecisive, gives the first place to C, the second to A, and the third to B. Manuscripts used for Book I.I will now demonstrate that the same conclusion is reached by following another line of investigation. Let us examine the various readings in the three MSS. and seewhether, in some instances at least, it is not possible to discover the causes which produced them. If it can be shown that in A and B there are a large number of readings which have been altered in a particular way for a definite reason, and if the originals of these readings are found in C, the inference is obvious; and we may then presume that in other variations of reading, where no such reason is apparent, C also possesses the greater authority.
The argument depends on considerations which have been suggested by a careful study of the prosody of the Mathnawí. I need not go deeply into this on the present occasion, but as some knowledge of the subject is essential even to those who read the poem in a critical text, the following remarks will not be out of place.
Jaláluʾddín's versification is lax compared with that of poets like Saʿdí and Ḥáfidh. The iẓáfat is frequently dropped after {{{1}}} (

- ↑ In the oldest MSS. and {{{1}}} are sometimes written {{{1}}} and {{{1}}}, even when they must be scanned as short syllables before consonants.