MARCH, 1873.]

CORRESPONDENCE, &c.

sūtra tells us that the termination ka applied to the names of objects, in the sense of images of those objects, is dropped in cases when the images enable one to earn his livelihood, but are not saleable. Upon this Patanjali observes that, because the word unsaleable' is used, such forms as Sirah, Skandah, and Visakhah (in which the termination ka is

dropped) are not valid. Why not? Because the Mauryas, desirous of obtaining gold, used, or applied to their purpose, i.e., sold, objects” of wor ship. Since, then, these (riz., images of Siva, &c.) were sold by them, they were panya, or ‘saleable,” and hence the termination ka should not be drop

ped. It may not be dropped in those cases (i.e., the proper forms must be Sivaka, &c.), says Patanjali, but it is dropped in the case of those images which are now used for worship. This interpretation of the passage is consistent and proper. Prof. Weber understands it to mean, that the only cases in which the rule about the dropping of the termination does not apply, are those of images with which the Mauryas were concerned. But that it is inapplicable to all images that are saleable, is clear from the passage itself, and the two commentaries on it. Kaiyata distinctly says that the rule does not apply to those that are sold,

and gives Sivakån Vikrinite as an instance. What Patanjali means to say is that the termination ka should be applied to the names of the images sold by the Mauryas, according to Pānini's rule; but the rule is set aside in this case, and the wrong forms Siva, Skanda, and Visãkha are used. Nagoji bhatta expressly states—tatra pratyaya-sararanam ishtamereti radan sūtrasyodaharanam dars'ayati (i.e., saying that the use of the termination there is necessary, he points out an instance of the rule). Now, in all this there is not only nothing to show that Pánini had the images sold by the Mauryas in view, but that the names of those images violate his rule. Dr. Goldstücker's interpretation of this passage is also not correct. In the next place Prof. Weber thinks that the word āchārya in such expressions as pasſyati trácháryah, occurring in the Mahābhāshya, applies to Patanjali. It appears to me that Prof. Weber has overlooked the context of these passages. In all these cases the fichārya meant is clearly Pānini, and not Patanjali. I will here briefly examine two or three of the passages referred to by the Professor, for I have no space for more. In the first of these,

95

hence an signifies only the vowels a, i, and u. why is it to be so understood 2

And

The sūtra ur an

raparah means, when an is substituted for ri, it is always followed by r, that is, if, for instance, you are told in a siltra to substitute a for ri, you should substitute not a alone, but ar. Now, the reason why, in this siltra, an signifies the first three vowels only, is that there is no other significate of the more comprehensive term an, that is, no other vowel or any semi-vowel or h which is ever substituted

for ri. “Why not ? there is,” says the objector. One instance brought forward by him is explained away, and another that he adduces is Maſtrinam.

In this case, by the sūtra námi, a long vowel, i.e., ri, is substituted for the short ri. Ri is a significate of the more comprehensive an, and not of the less comprehensive. Hence, then, the objector would say the an, in the sūtra uran, &c., is the more com prehensive one. But, says the siddhánti, this is not a case in which the substitute has an r added on to it. Does it follow from Pānini's work itself

that no ris to be added ? For aught we know, Pānini may have meant that r should be added in this case also. Now, the evidence from Pānini for this is in the sūtra rita iddhātoh.

‘This is the reason,'

says the siddhánti, ‘why the word dhātu is put in the sūtra, -that in such cases as Mátrinam and Pitrinam, which are not dhātus, ir may not be sub stituted for the long ri. If the long vowel substi tute in Maitrinam had an r following it, it would not be necessary to put the word dhātu in this siltra, for Mátrir would not then be an anga or basef ending in ri, and such bases only are intended in the sūtra rita iddhātoh.

The

use of the word

dhātu then shows that “the āchārya sees that in Matrinam, &c., the long substitute has not an r following it, and hence he uses the word dhātu in the sūtra.” + Now, it is evident from this that the āchārya is Pānini, for the āchārya is spoken of as having put the word dhātu in the sūtra for a certain purpose. The author of the siltras being Pånini, the āchārya meant must be he himself. In the same manner, in the passage at page 196 (Ballant. edition), Pānini is intended, for the fichārya is there spoken of as having put t after ri in the sūtra

urrit. Similarly, in page 197, the fichārya is repre

of the siltra a, i, un, and

sented as having used n twice in the pratyāhāra sūtras. The author of these siltras, then, is meant there. And I may say that, so far as I have seen the Bhāshya, the word āchārya used in this way applies either to Pánini or Kātyāyana, and Patanjali never speaks of himself as āchārya. Thirdly.—Prof. Weber's interpretation of the vár tika parokshe cha loka, &c., is different from Dr. Gold sticker's and mine. But he will see that our interpre tation is confirmed by Kaiyata and Nāgojibhatta. He seems to take paroksham in the sense of the past.’

• The reading in the Banaras edition is archyāh, and not

wrong or not good. It ought to be anantyatvad, as in the

the question Patanjali discusses is this:—Which n is it that is used in the term an occurring in the

sūtra ur an raparah, i. e., does an h

In ean

only a, i, and u, or all the vowels, semi-vowels, and h? He answers by saying that the n in this case is

clearly the first, and not the second, that is, that which is at the end

new Banaras edition. archáh.

f Antyatvád, the reading in Ballantyne's Mahābhāshya, is

I Pasyatitvácharyo nätra raparatvam bhavati tato dhā tu agrahann karoti.