Pramāṇa · Search · About · API

The Purana Text of the Dynasties of the Kali Age c7: Appendix II. The Oldest Scripts used in the Account

The Purāṇa Text of the Dynasties of the Kali Age c7

Language
English

Appendix II. The Oldest Scripts used in the Account

c7.2

Mistakes are found in the MSS, which can, it seems, be only explained satisfactorily by supposing that they arose out of misreadings of the ancient scripts (see Introdn. § 26). Some mistakes are obviously mere clerical blunders, but others cannot be accounted for naturally in that way. Kharoṣṭhī being the oldest Indian script that we know of, mistakes that could be traced to misreadings of its letters would be most significant. Such instances may singly be open to some distrust, but collectively they would have cumulative force; and without pronouncing a positive opinion, it does yet seem to me that certain misreadings do point to Kharoṣṭhī as their source. Such mistakes may prevail in many MSS, if they passed undetected from the beginning; otherwise they may only occur in single MSS, having been corrected in all the others.

c7.3

i. First may be cited an instance from the Vṣ, because it affords the best illustration of a misreading that seems significant, though the Vṣ does not contain the oldest version. It calls Aśoka generally Aśokavardhana, but kVṣ has Ayośokavardhana (p. 28, note 28). Here *yo* is obviously a misreading of *śo*; the copyist read the *śo* as *yo* and wrote *yo*, then he (or some one else) perceived the mistake and wrote or inserted *śo* in the copy, but the *yo* was not cancelled and the erroneous name Ayośoka remained and was repeated till it appears in kVṣ. Now *śo* could not be mistakenly read as *yo* in any Indian script except Kharoṣṭhī, and in that *śo* and *yo* were often written so much alike, that it is very difficult at times to say merely from the shape which letter was meant. Hence it seems reasonably certain that this passage in the Viṣṇu must have been originally taken from a Kharoṣṭhī MS. Had this mistake occurred in verse, the extra syllable would probably have been detected and the error corrected, but there was no such check in the prose of the Vṣ, and the mistake might have been followed in one copy (from which was descended kVṣ) though rectified in others.

c7.4

Other misreadings of *ś* and *y* occur, namely—*Ayoda* for *Aśoka* in *lMt* 1, where the second misreading of *k* as *d* might have arisen later in the Gupta script 2; *Māgadhéso* in *jMt* 3 where the more general readings are *Māgadhā ye*, *Māgadho yo* or *Māgadheya*; *Koyāla* in cgVṣ 4 for *Kośala*, where *yā* might easily be read for *śa* because Kharoṣṭhī often did not distinguish between long and short vowels; and *Śalīyūka* in eVā for *Śalīsūka* 5; *Mauryā dayo daśa* in *lVṣ* 6, where *daśa* was probably first misread and written as *daya*, which was afterwards amended so as to read *Maury-ādayo* incorrectly.

c7.5

ii. Some similar variations seem to point to the same conclusion. The Mt reading, Kāśeyās, appears in djMt as Kāleyās; and the mistake of l for ś seems best explainable by their similarity in Kharoṣṭhī. The Vā and Bd read Kālakās, which is probably a similar misreading of the equivalent name Kāśakās¹.

c7.6

¹ P. 27, note 3.

c7.7

² See Bühler's Ind. Palaeog., Table IV, cols. xxi, xxiii, and Table V, cols. viii, ix.

c7.8

³ P. 14, note 1.

c7.9

⁴ P. 54, note 20. *Koyāla* is an error in writing, different from *Kojala* which was a variation of *Kausalya* in pronunciation; see Actes du XIVe Congrès International des Orientalistes, Alger, 1905, p. 217.

c7.10

⁵ P. 29, note 37.

c7.11

⁶ P. 30, note 46.

c7.12

¹ P. 23, note 9.

c7.13

iii. Two other letters which might be confused in Kharoṣṭhī but not in any other script are k and bh, and there are some variations which seem to have so originated. The Vā generally, and the Mt sometimes, have Tuṅga instead of Śuṅga², a misreading the cause of which is not clear³, but the name Śuṅgabhṛtya is generally corrupted to Tuṅgakṛtya in the Vā, while eVā alone among the Vā MSS has preserved it nearly right as Śuṅgavṛtya⁴. Here it seems certain that bh was misread as k in a Kharoṣṭhī MS. The converse appears to be the cause of the faulty Mt reading in p. 41, line 22, where saumyo bhaviṣyati, with no mention of the length of the reign, seems to be a misreading of the Vā and Bd reading so 'py eka-viṃśatim, for, while saumyo might be a later mistake and emendation for sopye, bhaviṣyati could be a misreading of ka-viṃśati in Kharoṣṭhī only. The two forms would be bhavissati and ka-visati in Pali and probably also in literary Pkt, and these two would be almost identical in Kharoṣṭhī which generally wrote long and short vowels alike and doubled letters as single.

c7.14

² P. 32, note 47; p. 33, note 52.

c7.15

³ Perhaps through the Pkt form Śuṅga; s carelessly made might be read as t in Kharoṣṭhī. The mistake is ancient as it is found so widely.

c7.16

⁴ P. 34, note 25; vṛtya might be a modern misreading of kṛtya.

c7.17

iv. As regards Brāhmī, I have not found any variations of importance which can be assigned definitely to misreadings of it, and there is not the same scope for detecting such errors, because there is more resemblance between Brāhmī and Gupta letters than between them and Kharoṣṭhī. All the mistakes that I have detected, which might be attributed to misreadings of Brāhmī letters, might equally well, or even better, be attributed to misreadings of Gupta letters. Hence it seems to me, speaking with diffidence, that no light is thrown by Brāhmī on the age of the account or the MSS, and that, so far as the negative argument is of weight, Brāhmī writing played no part in the early MSS of these dynastic accounts. If this be so, the accounts passed from Kharoṣṭhī into the Gupta script.

c7.18

v. If these explanations of these variations be reasonable and not fanciful, it appears that the Mt, Vā, and Vṣ all betray the fact that their accounts were originally copied from MSS written in Kharoṣṭhī. This script was in use till A.D. 300, or perhaps even half a century later⁵. This conclusion would, as regards the Mt and Vā, agree with the period assigned to them⁶. There is no further indication regarding the date of the Vṣ, and as Kharoṣṭhī MSS would have lasted some centuries, the Vṣ account might well be later and yet have been extracted from such a MS. There has been no opportunity of testing the Bd account in this way, because I have not been able to collate any MS of it; and the printed edition betrays no misreadings of this kind; but it is so closely like the Vā that the same conclusion probably holds good for it.

c7.19

⁵ JRAS, 1907, pp. 184-5.

c7.20

⁶ See Introdn. §§ 21-24.

c7.21

vi. Nor have I found any variations in the Bhāgavata which point to misreadings of Kharoṣṭhī or even of Brāhmī. I have noticed only two peculiarities which may perhaps be significant.

c7.22

In the list of Andhra kings Hāla was succeeded by a king whose name consisted of four syllables, the best supported forms of which are Mantalaka or Pattalaka¹. The Bh calls them *Hāleya* and *Talaka* respectively, *Hāleya* ending the first half of a line and *Talaka* beginning the second half, thus:—

c7.23

Aniṣṭakarmā Hāleyas Talakas tasya cātmajaḥ.

c7.24

These two names seem to be mistakes for *Hāla* and *Pattalaka*, the *pa* being misread as *ya*. If so, the wrong division of these two names in the middle of a line seems only explicable on the supposition that the Bh prepared this verse from a bare prose list of kings and divided the letters of the two names incorrectly. If this suggestion has any validity, it would appear that the Bh could not have been composed till after the time when *y* approximated to *p* in shape, that is, after the 7th century A.D.

c7.25

The other instance is the name of the Andhra king Āpilaka, which appears in the Bh generally as *Civilaka*. The probable genesis of the changes in the name is suggested in p. 39, note 45, and the fact that seems significant here is that the compiler of the Bh account apparently drew his information from a Vs account in which he misread the initial *d* as *c*. This mistake could arise only in the Gupta script and not very well before the 7th century A.D.²

c7.26

² See Bühler's Ind. Pal., Table IV.

Linked texts (5)