________________
228
EPIGRAPHIA INDICA
[VOL. XXX
Next we come to the word Saramgha. This is also of non-Sanskritic origin and is obviously connected with the Middle and New Persian Sar-hang meaning 'head of the army of gathering'. Its origin has to be sought in the Iranian sar, 'head' (Indo-Aryan siras) and the old Persian hanga (8kt. samgha), company'. Even now the Indianised form of the word can be traced in the term Sareng, head of the crew of a steamer'. The term Käñchudi, as was pointed out to me by Sir Aurel, must be connected with the racial designation of Kanjuti applied to the people of Hunza and known as Kanjut to its neighbours.
According to the inscription, king Patōladeva was born in the lineage of Bhagadatta who is no other than the homonymous son of Naraka mentioned in the Mahabharata, It is interesting to note that the same lineage is claimed also by Bhaskaravarman, the ruler of Prägjyotisha (Assam) and the contemporary of Harsha of the Pushabhūti family, ruler of Kanauj and Thanesvar (7th century A.D.). How the rulers of two widely separated territories, one in the mountainous region of the north and the other in the extreme east, came to trace their descent from the same ancestor, it is difficult to explain. It may be that both had the same object in view, namely, to establish their origin to a reputed Kshatriya family stated to be descented from the god Vishnu himself. Of course the name of Prägjyotisha was well known in Kashmir in ancient times. Kalhana refers to it on three occasions, once in connection with a story in the Mahabharata and twice with the kings of Kashmir. In Book II (vv. 146 ff.) it is stated that Meghavahana, who became the king of Kashmir on the restoration of the Gōnanda dynasty, won the hands of Amritaprabha, daughter of the king of Prägjyotisha, in a svayamvara ceremony. There is also a reference to the Assian kings' descent from Vishnu and the parasol of Varuna which was carried there by Naraka. Kalhana also mentions this country in connection with the digvijaya of Muktapida Lalitäditya (8th century). But what is strange is that immediately after the territories of the Bhuttas and Daradas, he mentions Prägjyotisha to be followed by only mythical regions in the north. In the first instance also, while Amritaparbha is mentioned as a princess of Pragjyotisha, her father had a guru who was obviously a Tibetan. Can these instances indicate that there existed a tradition in Kashmir of a second Prägjyotisha in the north of Kashmir in the neighbourhood of the Darada country? Or, was it that the kings of Pragjyotisha in Eastern India were in some way connected with the region in the north of Kashmir? If we can trace such a tradition that would offer an easy explanation for connecting the family of Patōladeva with Bhagadatta. The kingdom of the Assam rulers might have extended to a part of the hills but not certainly so far to the west.
Another point is that the Gilgit area is immediately across the Hindukush adjoined by Iranian territory and Stein has pointed out that in Wakhan the epic tradition of Iran was fully alive among the people. He also informed me that, even in the south, the Ishkuman valley is partly occupied by modern immigrants from Wakhan, speaking an Eastern Iranian tongue, and its present ruling family came from there. Thus it is not unlikely that, in an earlier period, to the Indianised descendants of the Iranian Kushāņas a derivation of their traditional family claim from a legendary hero of the Mahabharata might well have appealed. But in the absenceof historical records nothing can be established. The Chinese sources do not help us much as Chinese authority over these parts ended much earlier and after the Islamisation of the territory all such traditions seem to have been altogether lost.
Following, the discovery in 1931 of the now wellknown Buddhist Mss. in a stupa in the mountainous region 3 miles to the north of Gilgit," Pandit Madhusudan Kaul of the Kashmir
1 See Yule, Hobson Jobson, s.v. Serang.
M.A. Stein, Chronicles of Kashmir, Vol. II (Bk. VIII, v. 2811).
Ibid., Vol. I, p. 137 (Bk. IV, v. 171).
Ibid., Vol. I, p. 73 (Bk. III, v. 10).
Hackin, Journal Asiatique, 1932, pp. 14-15.
