Historical Documents of Kulu
This is a reproduction of the paper “Historical Documents of Kulu” by Hiranand Shastri published in the Annual Report (1911) of Archaeological Survey of India for the year 1907-08.
KULU is a sub-division of the Kangra district of the Panjab and comprises the Upper Bias Valley, Waziri Rupi, Saraj, Lahul and Spiti. It lies between 31° 20′ and 30° 55′ latitude and 76° 50′ and 78° 35″ longitude. On the north it is bounded by Ladakh, on the east by Tibet, on the west by the principalities of Chamba, Mandi and Suket, on the south by the Satluj and the Bushahr States. The total area of the sub-division is 6,025 square miles, which is a little less than half the area of Holland. Its population, according to the census of 1901, is only 19,585, which is about one-fifth of that of Amsterdam. For administrative purposes this tract is now divided into two tahsils, called Kulu and Saraj. The latter and a portion of the former tahsil together form a homogeneous region which may be termed Kulu proper. The remainder of the Kulu tahsil consists of Lahul and Spiti.
As has been pointed out by General Cunningham, the old name of Kulu was Kuluta, a term which from Hiuen Tsiang’s account, noticed below, appears to have designated Kulu proper. According to a popular derivation, which is also admitted by Captain Harcourt, the valley was originally called Kulantapitha, signifying the end of the inhabitable world, as the Hindus considered it to be the last boundaries of civilization. The name, indeed, occurs in a booklet called Kulantapithamahatmya, which will be noticed subsequently. Phonetically the change of Kulantapitha into Kulu is an impossibility, and the derivation should, therefore, be treated as an instance of popular etymology. Some connect the name of Kulu with Sanskrit Kaula meaning devi-worshipper. Others trace it to the caste names Kol and Koli. These derivations are evidently fanciful and must be at once set aside.
The material for building up the history of Kulu is very scanty. This scarcity is perhaps due to the proverbial ignorance of the people. References scattered in Sanskrit literature and a few epigraphical and other documents that have come to light are noticed below, with a view to glean from them facts regarding the history of the valley.
The Mahabharata mentions Kuluta in the list of countries in the north of India. The Markandeya-purana and the Brihtasamhita notice it among the tracts situated in the north-east of India. Obviously then Kuluta had a distinct existence in old times. Kalhana indicates that about the 6th century it was a separate State, when he says that Ratisena, the king of the Cholas, sent his daughter Ranarambha to the residence of his friend, the king of Kuluta, and “Ranaditya went with joy to that not distant land to receive her.” Bana (middle of the 7th century A.D.) in his Kadambari tells us that Kuluta was conquered by the great King Tarapida of Ujjayini, who took captive the princess Patralekha, the daughter of the king of that country, and that Queen Vilasvati sent her to Prince Chandrapida, her son, to be his betel-bearer. Tarapida of Ujjayini is not known to history, but it is curious that Chandrapida and Tarapida are the names of the immediate predecessors of Lalitaditya-Muktapida of Kasmir. At any rate it shows that Bana recognized Kulu as a distinct kingdom. From the chronicle of Jonaraja we find that in the fifteenth century the valley was in the possession of a Tibetan power. For it is stated there that Sultan Zainu-l-‘abidin (A.D. 1420-70) in an expedition against Goggadesa, i.e., the kingdom of Guge, robbed by his splendour the glory of the town of Kuluta.
Hiuen Tsiang places the country of K’iu-lu-to at 700 li, i.e., 117 miles to the north-east of Jalandhara. This exactly corresponds with the position of Kulu which must be identical with Kiu-lu-to — the Chinese rendering of the Sanskrit Kuluta. The circuit of this tract given by the pilgrim is 3,000 li, i.e., 500 miles. This figure is very much exaggerated, if compared with the present limits of Kulu, and might have been given on hearsay. Besides, we know that Hiuen Tsiang usually overrates distances in mountainous tracts. For instance, he places the country of Lo-u-lo (Lahul) about 1,800 or 1,900 li to the north of the country of Kiu-lu-to (Kulu), though the distance between the capital of these two countries, i.e., between Nagar and Kyelang, is only about 400 li or 70 miles approximately. Cunningham, however, relies upon tradition and is inclined to accept the above figure. For he says: “As the ancient kingdom is said by the people themselves to have Included Mandi and Suket on the west, and a large tract of territory to the south of the Satluj, it is probable that the frontier measurement of 500 miles may be very near the truth, if taken in road distance.’’ But a more common tradition limits the ancient territory of the State to Waziri Parol which was extended in the reign of Bahadur Singh in the 16th century. The account given by the pilgrim shows that in the beginning of the 7th century Kulu formed a distinct State. He makes mention of a Stupa, which Asoka had built in the middle of the country to commemorate Buddha’s visit to the valley. He further tells us that in his time there were about twenty sangharamas and some 3,000 priests, who mostly studied the Great Vehicle. There were moreover fifteen Deva temples which different sects used without distinction. Along the precipitous sides of the mountains were found stone chambers facing one another and hollowed in the rocks, wherein the Arhats dwelled and the Rishis stayed. This shows how flourishing Buddhism must have been here in the 7th century, though it has now practically disappeared from the valley. A stone image of Avalokitesvara is still being worshipped at Kelat near Sultanpur, but the people call it Kapilamuni. The description given of the climate, the fauna and flora, of course, holds good now also.
Among the local records that have been brought to light the Bansauli, or genealogical list on which Captain Harcourt based his account of the rulers of Kulu, stands foremost in virtue of the information it supplies. It is much to be regretted that the original document was not published, for it is not forthcoming now. Such records, provided they are genuine, are very interesting and possess great historical value. Copying might invalidate them, yet much of the information they supply is often confirmed by other sources. The Chamba Bansauli, for instance, gives, as is proved by copper-plate inscriptions, the correct names of the rajas of Chamba after A.D. 1300. The names of those who ruled between A.D. 700 and 1300 are partly changed; but there the person who copied the old records may be at fault.
The list, which the Bansauli of Kulu contains, is also corroborated by copper-plate and other inscriptions, so far as the Singh dynasty is concerned. For example, the grant of Bahadur Singh mentions as second donor the heir-apparent Pratap Singh, whom the roll rightly makes the successor of the former. The Tibetan records also confirm it. It is true that such local chronicles are apt to exaggerate the exploits of the families to which they belong and leave out events which detract from their glory. The Bansauli of Chamba relates, for instance, that Raja Prithi Singh conquered Jagat Singh Pathaniya of Nurpur, while we know from other sources that he played only a secondary part there. The Kulu chronicle makes no mention of the fact that Kulu ever fell under the sway of Ladakh or was conquered by Zainu-l-‘abidin Accordingly, the accounts which these records contain should be sifted and compared with those supplied by other sources. Still for all that, it cannot be denied that they possess much historical value.
The roll published in Captain Harcourt’s book (pp. 370-5) gives a list of no less than eighty-eight rulers of Kulu. The names of the earlier princes end in Pal starting from Bihangamani(?) Pal, the reputed founder of the dynasty Sidh Pal, 74th in descent from the first Pal, was the first Raja who took the surname of Singh, and since his time the chiefs of Kulu have had names ending in Singh. Bihangamani and his eleven immediate successors reigned at Jagat Sukh, the old capital of Kulu. In the reign of Uttam Pal, the twelfth prince, the seat of Government was transferred to Nagar, whence it was finally shifted to Sultanpur in the reign of Jagat Singh who was eightieth in descent from the founder of the dynasty. It is said that during the rule of Sridattesvar Pal, the 31st Pal, there was war with Chamba, in which Amar Singh, the ruler of the latter country, who was contemporary with a Delhi (Indrapat) king, Gobardhan, killed the Kulu Raja. This statement is obviously wrong, for no such ruler of Chamba is known. The names of the chiefs of this country, ended in varman down to the time of Ganesavarman whose successor Pratap Singh was the first Raja of Chamba who adopted the cognomen Singh. May not Amar Singh be the name of some petty chief or commander under the ruler of Chamba?
During the time that followed, Chamba must have held sway over Kulu, till Srijaresvar (?) Pal asserted his independence. In the reign of Narad Pal, who was fortieth in descent from Bihangamani, war again broke out with Chamba and lasted for twelve years. The troops of Chamba advanced as far as Majnakot, a village near the Rohtang pass, and built a fort there. But a peace was patched up at last and the people of Kulu, inveigling the soldiers of Chamba, threw them into the Bias near Ralla, where they all perished. The Bansauli further informs us that Sikandar Pal, the fiftieth Pal, went to the king of Delhi to seek shelter against the Chinese who had invaded his kingdom. The Raja of Delhi headed an army in person and, marching through Kulu, took Gya Murr Orr (?) and Baltistan, together with the country as far as Mantilae, i.e., Mansarovar lake. This, if true, is a curious record, for it shows, as Captain Harcourt remarks, that an Indian army could successfully penetrate so far north as the Mansarovar lake. During the reign of Nirati Pal, Ali Sher Khan is said to have ruled in Kasmir. No ruler of Kasmir had such a name, but it is possible that Ali Shah (A.D. 1413-20), the brother and predecessor of Zainu-l-‘abidin, is meant. Kulu was conquered by Bushahr, Kangra and Suket during the reign of Hast Pal, Nand Pal and Kirat Pal, who were respectively 56th, 62nd and 67th in descent from their progenitor.
The kingdom of Kulu was consolidated and enlarged in the time of Bahadur Singh, who was the 75th raja of the valley. Jagat Singh, the fifth successor of this prince, had a long reign of sixty-one years, during which he considerably extended his dominions by conquest. His rule was contemporary with the latter part of Shah Jahan’s and the earlier years of Aurangzeb’s reign. He introduced the worship of Vishnu in the form of Rama and Krishna into the valley. Bidhi Singh, his successor, made Kulu a really important state. All Kulu, Waziri Rupi, Saraj, Lahul, Spiti, Bushahr, Bangahal, a great portion of Mandi and Suket, with the hill states close up to Simla were under this chief’s sway. Man Singh, who succeeded him, subdued Mandi entirely, but gave it up afterwards. It was in his time that the power of Kulu reached its zenith. The princes who followed this powerful Raja could not keep up their position. Jay Singh, his second successor, had to go to Lahore and ask for aid against Mandi. He was in turn succeeded by Tedhi (“the Crooked”) Singh and Pritam Singh. During the rule of the latter the country was torm by internal dissensions. In Bikram Singh’s time the Sikhs invaded Kulu. Jit Singh, the last ruling chief, who was 88th in descent from the founder of the dynasty, was deposed in A.D. 1840. With this event the old principality of Kuluta ceased to exist. In 1841 Jit Singh died without issue at Shangri, where he had retired after having escaped the oppression of the Sikhs. Jhagar Singh, his uncle, was with him at the time of his death and got possession of Shangri. Rai Hira Singh, Jhagar Singh’s son, still holds his Jagir at that place. Thakur Singh, a collateral of Jit Singh, was made titular raja, and Waziri Rupi was given to him in jagir. Gyan Singh, his son and successor, was called Rai instead of Raja. Megh Singh, the present Rai of Rupi, is Gyan Singh’s grandson.
We have already noticed that there exists a booklet bearing on the sacred ore of the valley, which, though not published, can be seen with the priests of Manikarn, the principal place of pilgrimage in Kulu. It is called Kulantapitha-mahatmya and pretends to be a part of the Brahmanda-purana. Though possessing little historical interest, it is not unimportant for local topography and its contents maybe noticed here briefly. “Kulantapitha,” it says, lies to the north-east of Jalandhara and south of Hemakuta mountain. It is ten yojanas (about 90 miles) in length and three (?) yojanas (about 27 miles) in width. The sacred place of Vyasa lies to its north and the Bandhana mountain to its south. The river Bias flows to its west and the Paśupati (Śiva) lies to the west. The deity presiding over the valley is Śavari. Indrakila is the principal hill. The Samgama or confluence of the Bias and the Parbati river is the chief sacred place. It was in this land that Śiva in the guise of a Śavara fought with Arjuna.
Though the extent which this Mahatmya gives to the Kulantapitha is nearly equal to that of Kulu proper, yet there is hardly any reason to assume that the latter term ever meant Kulu. Mr. G. C. L. Howell, Assistant Commissioner of Kulu, informs me that this designation is still applied to the tract on the left bank of the Bias, between the Bias-Parbati confluence and the source of the Bias, the latter being its western boundary. This then is another argument against the assumption that the appellation Kulantapitha dwindled into Kulu in course of years. The information supplied by the above named officer enables me to identify some of the names mentioned in the text. The northern limit of the tract (pitha) is termed Hemakuta, which according to the Puranas is a Simaparvata, i.e., a boundary mountain. The Pir Panjal range being the northern boundary of Kulu, Hemakuta, if it means any particular peak, would mean the Snowy Peak M. of the said range, wherein lies the true source of the Bias, known as Bias Kundi, the place with which the Vyasatirtha of the text must be identified. Indrakila is the name of the well-known mountain in the same range and lies to the south of the Hamta. It must be nearly 20,000 feet high and in shape resembles a wedge, whence the designation kila. Both these features give it a very impressive appearance, in consequence of which perhaps it is so well-known in the Puranic literature.
Another important source for the history of Kulu is the Tibetan chronicle of Ladakh rGyal-rabs or “the Book of Kings”. What we gather from it is this: Skyid-lde-nyima-mgon (about A.D. 1000), the first king of West Tibet or Ladakh had three sons, of whom Lha-chin-dpal-gyi-mgon was the eldest. The latter had two sons aGro-mgon and Chos-mgon. The great-grandson of aGro-mgon was Lha-chen-rgyalpo, who, according to Mr. Francke reigned between A.D. 1030 and 1080.
His son was Lha-chen-Utpala (A.D. 1080-1110), who united the forces of Upper and Lower Ladakh and invaded Nyungti or Kulu. The ruler of the latter country bound himself by oath “so long as the glaciers of the Kailas will not melt away, or Mansarovar lake dry up” to pay his tribute in Dzos and iron to the king of Ladakh. This treaty remained in force at least down to the time of king Sengge-namgyal, and Dr. Marx tells us that the tax collectors of the king of Ladakh used to visit Lahul and probably Kulu till A.D. 1870, although the two districts were then already under British rule. Further on we find that Tsewang rNamgyal I (These dbang) between A.D. 1530 and 1560 subdued Kulu “whose chiefs were made to feel the weight of his arm.” From the high titles assumed by Bahadur Singh in his copper-plate grant, it would appear that this event happened before the latter prince rose to power in Kulu; these titles paramabhattaraka maharajadhiraja would suit an independent raja only. Further, in the beginning of the nineteenth century (i.e., when Bikram or Vikramajit Singh ruled over Kulu and Tsepal over Ladakh) we find that, encouraged by the gross carelessness of the latter prince, the army of Kulu invaded Spiti and, after having destroyed the villages, carried away all property. Later on, the people of Kulu and Lahul combined against Zangskar, laid waste that tract and took away whatever was valuable.
The inscriptions, hitherto discovered, which throw considerable light on the history of the valley, may now be briefly noticed. The foremost of these is the legend on the coin of a Kuluta king Virayaśa, which reads rajna Kolutasya Viryaśasya. As has been pointed out by Professor Rapson, it can be ascribed on palaeographical grounds to the first or second century A.D. Here, then, we have the earliest archaeological record of the Kulutas. Next in date is the rock inscription at Salri near Salanu, which though lying in the territory of Mandi, may be regarded as geographically belonging to the Kulu valley. The characters of this record are of the fourth or fifth century A.D. Its purport is to record that a Maharaja Śri-Chandeśvarahastin, who was the son of a Maharaja Iśvarahastin and belonged to the family of Vatsa (?), conquered in battle a Rajjilabala (?) and founded a town of which the name apparently was Salipuri. This town, I think, is the present Salri village situated near the inscription. It is not known who these personages were, and consequently the significance of the document (Plate LXXXIII) cannot at present be fully realized.
From the copper-plate grants of Rajas Somavarman and Asata of Chamba we learn that in the eleventh century the dynasties of Chamba and Kulu were related, and allied to each other. They describe the Chamba raja Sahilla as one “who was asked the favour of bestowing royalty in return for services by his kinsman the Lord of Kuluta anxious to render him homage.” Here the use of the attribute “kinsman” (svakulya) calls for remark. The ruling family of Chamba is Kshatriya by caste and consequently the Lords of Kuluta must also have been Kshatriyas. Otherwise they could not be called svakulya meaning “of one’s own family or relative.” This inference is confirmed by Bahadur Singh’s grant, which mentions three Kulu princesses given in marriage apparently to the heir-apparent of Chamba. I may note here that Viśakhadatta classes the Kulutas with Mlechchhas. In the play Mudrarakshasa he mentions Chitravarman, the king of Kuluta, among the five leading Mlechchha allies of Rakshasa. Neither Chitravarman, nor the other confederate rajas, appear to have been historical persons. At any rate, this shows that in Viśakhadatta’s time (c. A.D. 600 ?) the people of Kulu were regarded as barbarians, if not foreigners. On the other hand, their coin noticed above proves for certain that the Kulutas had Indo-Aryan names in the first or second century A.D. Consequently, if they were non-Hindus, they must have rapidly become Hinduized like the Kshatrapas of Surashtra. In Visakhadatta’s time their origin could not but have been forgotten. That the latter should call them Mlechchhas is, therefore, a puzzle, if this statement is not to be regarded as a mere poetic licence. Or should we take svakulya in the sense of “well disposed or friendly to one’s family” and translate the expression svakulya-Kuluteśvara in the grants by “his friend the Lord of Kuluta”?
The epigraph found in the temple of Sandhya Devi at Jagat Sukh is another document of Kulu history. It is cut on two slabs placed on the enclosing wall at the entrance of the courtyard. It contains the name of Maharaja Udhran Pal and, the date 4 (?) ba. ti. 2 which corresponds with A.D. 1428. Evidently, this is the date of the temple as well as of Udhran Pal, the third predecessor of Bahadur Singh. While showing the importance of Jagatsukh long after it had ceased to be the capital, the record makes the shrine of Sandhya Devi the oldest temple in Kulu of which the date is known.
The inscription on the famous temple of Dhungri near Manali records the foundation of that sanctuary by Bahadur Singh in the Śastra year 29, i.e., A.D. 1553. It may be remarked here in passing that this shrine, where human sacrifices used to be made within the living memory of the people, possesses perhaps the finest specimens of wood carving in the valley (cf. fig. I).
Another important inscription belonging to this period, that has already been published by Dr. Vogel, is a grant of Bahadur Singh in favour of Ramapati, the Rajaguru of Chamba. From this we learn that the said ruler in A.D. 1559 governed the whole of Kulu proper and that the principality of Lag, situated between Bajaura and Dhungri, both of which places formed part of Bahadur Singh’s dominion, was tributary to Kulu. In this document the chief of Kulu is called Surātrāna (Suratrana) raja, i.e., “Raja Sultan”. The latter epithet has been connected with the name of Sultanpur, the present capital of Kulu, the origin of which it will not be out of place to notice here. According to tradition, preserved in the Bansauli, Jagatsukh was the first capital of Kulu. It is said that Bihangamani Pal, a fugitive prince from Mayapuri near Badri-Narayan, took shelter with a potter at Jagatsukh, and the people, disgusted with the rule of their Thakur, made him raja. A rock known by the name of Jagatipat, is still pointed out midway between Manali and Jagatsukh, where this prince used to sleep. He is regarded as the founder of the Kulu dynasty. The only relics of the former prosperity of the town are a few old sculptures placed in the shrine of Sandhya Devi, the best of which representing Ganga on her vehicle the makara, I lately secured for and deposited in the Lahore Museum (fig. 2).
Nagar is said to be the second capital, from where in the seventeenth century the seat of government was transferred to Sultanpur. Tibetan sources, however, take no notice of the first two capitals of Kulu. The chronicle of Tinan, for instance, completed in the reign of Bahadur Singh of Kulu, speaks of the latter ruler as residing in Magarsa. So do the documents belonging to the time of Partap Singh and Parbat Singh. It is under Pritam Singh, who ruled about A.D. 1780, that Sultanpur is first mentioned by the Tibetans under the form Setanpur. It is certain that, before the seat of government was shifted to Sultanpur, the capital of the valley was Nagar. How then could the Tibetans ignore the real capital and mention another instead? Perhaps they took Magarsa and Nagar as identical, i.e., they called Nagar by the name of Magarsa. Captain Harcourt identifies one with the other and supports his identification by the following statement of Moorcroft. “On the 11th we passed a house belonging to the Raja on our right, situated on an eminence, at the foot of which stood the ancient capital of Kulu called Makarsa. A few houses are all that remains of it, as the removal to Sultanpur took place about “three centuries ago.” Tradition, however, does not corroborate this identification. According to it, Magarsa is not a town or city but a tract or district of Kulu named after a town Makarahar which was founded by Makarsa, a son of Vidura of the Mahabharata. The site of this town is still pointed out near Hurla in Kothi Kot Kandi, some 22 miles north-west of Nagar. This tradition is preserved in the following popular saying:—
Rane Thakar marie keru bhurasa Makarahar basie raj banu Makarsa
“The Ranas and Thakurs were killed and smashed [by the rulers of Kulu]. Owing to the residence at Makarahar, the raj (State) became known by the name of Makarsa.”
Sardar Hardyal Singh in his account of Kulu states that Bahadur Singh, after completely subjugating the tract called Rupi, repopulated the ruined town of Makarahar, where he built a palace for his residence and died in A.D. 1569. It would have been interesting had that author given some authority for his assertion.
Rai Hira Singh of Shangri informs me that this town was very prosperous in the days of Jagat Singh, who erected there a temple of Rama and deposited in it a murali (flute) which he had obtained from Ayodhya. In support of this statement he sent me this couplet:—
Makarahar Ajodhyapuri mano hem Braj ki rit
Jagat Singh maharaj ki Sri Ragho-ji sem prit.
“Makarahar is another Ayodhya and is the counterpart of Braj (tract round Mathura). Maharaja Jagat Singh is devoted to the illustrious Ragho-ji (i.e., Ramachandra).”
Magarsa of the Tibetan chronicles would then be not a town but a tract or district where the above-named rulers resided. According to the genealogical roll, as has been remarked already, the capital was transferred from Nagar to Sultanpur in the days of Jagat Singh, who flourished at the close of Shah Jahan’s reign. Tradition ascribes the origin of the name of Sultanpur to one Sultan Chand, the brother of Raja Jog Chand of Lag. At the death of the latter, it is said, Jagat Singh (after killing Sultan Chand, who governed the tract round Sultanpur which he designated after himself) seized the whole country up the Sarvari valley as far as the Bhubu Pass. On account of its better situation, Jagat Singh preferred Sultanpur to Nagar and made it his capital. It was, as Captain Harcourt tells us, a regularly walled town, but now its fortifications have all been razed and there remain only two gateways on the north and south, both of which are difficult of access.
Dr. Vogel, however, holds that Sultanpur was probably founded by Bahadur Singh. He is led to this hypothesis by the title Sultan (Skr. Suratrana), which he takes to be a second name of this prince. But the grant, as far as I can see, mentions this as an epithet and not as a second name. Supposing Sultan Singh was another name, it is not clear why the primary appellation should not have been selected to designate the capital town; Bahadurpur would have been more appropriate. Nor is it apparent that importance can be attached to the circumstance adduced in support of this theory that “the name of the Raja of Lag, after whose death Jagat Singh annexed the territory, is neither Jay Chand nor Sultan Chand but Jog Chand” in the sanads to be noticed below. The foundation of Sultanpur is ascribed to Sultan Chand, and not to his brother. At any rate the roll and the chronicle of Ladakh, do not countenance that assumption, and there is little to commend it.
In the year 1904-05 the survey brought to light ten more lithic records in the valley. Five of these are dated between A.D. 1673 and 1870 and are partly illegible except one engraved on an image of Vishnu, at Hat near Bajaura, which, so far as decipherable, reads: Śri-Paramabhattaraka pa. They are all written in Tankari and composed in the local dialect. The most important of this lot are two. One written on the jambs of the doorway of the Śiva temple at Hat, is dated in the Śastra year 49 (A.D. 1673) and in the reign of Śyam Sen of Mandi, and records a grant of land to the temple. This indicates that in A.D. 1673 Hat was perhaps under the jurisdiction of Mandi. The second is on a stone slab in the wall of the Murlidhar temple at Chahni, two miles above Banjar (Inner Saraj). A part of it is written in Sanskrit. It was engraved in the reign of Bidhi Singh in Śri Samvat 50, on the 15th day of Pausha and in the Vikrama year 1731 (A.D. 1674-5). Four more inscriptions on copper-plate were discovered. Two belong to the reign of Jagat Singh, one being dated in the Śastra or Saptarshi year 27 (A.D. 1651) and the other in 32 (A.D. 1656). They record grants of land, and show that the worship of Vishnu in the form of Rama and Krishna, became the State religion in Kulu about the middle of the seventeenth century, i.e., in the time of Jagat Singh who consigned his State to Rama and acted as an agent of the Lord. The third record belongs to the reign of Raj Singh, but is not dated. The fourth was written in the Śastra year 56 (A.D. 1780) in the reign of Pritam Singh.
Besides the documents noticed above, the valley possesses epigraphical material of chronological interest in the inscriptions on metal masks called deo, which represent Hindu gods and deified personages. Unfortunately, the tendency to replace old objects by new ones has caused much loss, as the custodians unscrupulously melted the old masks in order to renew them. Perhaps it is owing to this circumstance that very few old inscribed masks are met with. Of the ancient Pal dynasty only two have been noticed. One, engraved on the mask of Hirma, gives 94 as the date for Udhran Pal. The other, on the effigy of Vishnu at Sajala in Kothi Barsaiya, gives Śastra-Samvat 76 and Saura year 1422 as the date of Sidh Pal. Evidently Saura stands for Śaka, and 1500 is the equivalent Christian date of Sidh Pal. This, therefore, shows that Udhran Pal, the second predecessor of Sidh Pal, flourished in the beginning of the fifteenth century A.D. Accordingly the date of the Sandhya Devi temple must be 1428 and not 1528, the alternative proposed by Dr. Vogel. As the rajas of the Singh dynasty are comparatively modern, their gifts are better preserved and consequently the names of most of them are found on these masks. Their dates are also given, but generally in the Śastra era. In none of them does the week day or any other chronological date seem to be indicated, excepting the pravishta, viz., the day of the month. These dates therefore do not admit of verification.
Some farmans or official letters issued from the Mughal Court at Delhi between the Hijri years 1061 and 1068 (A.D. 1650 and 1658) to Raja Jagat Singh of Kulu also furnish interesting information. These are thirteen in number, four of which are original sanads in possession of Rai Hira Singh of Shangri, while nine are copies, the originals of which are lost, belonging to the present Rai of Rupi. Twelve were issued under the seal of Dara Shikoh, and one in A. H. 1068 by Aurangzeb. In the latter, Jagat Singh is asked to join hands with Dhan Chand Kahluria of Bilaspur in order to close the roads against Sulaiman Shikoh, who desired to join his father at Lahore from Garhwal. Sulaiman Shikoh, as we learn from Bernier and Manucci, fled to the hills after his father Dara had lost the battle at Samu-garh. This battle, it should be remembered, took place in June 1658. Manucci wrongly gives 1656, as has been pointed out by Mr. Irvine. One of these sanads addressed to Sayyid Ibrahim, who appears to have been an officer of the court of Delhi, placed in charge of the Kangra valley. The prefix shows him to be a different person from the General Ibrahim Khan, who was with Dara Shikoh, at the battle near Samu-garh. The remaining farmans are addressed to Jagat Singh, who is therein called zamindar, i.e., “landlord” of Kulu. It is only once that he is styled Raja. What we gather from them is this: Jagat Singh was held in considerable esteem at the Mughal Court, for Aurangzeb spoke of him as “well-established in his royal ways”. He sent presents of hawks and crystal, and deputed his son to the Imperial Court at Delhi, and thus recognized Shah Jahan, as his liege lord, who in turn was highly pleased with him and, besides granting some crystal mines, extended his protection to him and his people. Later on, however, he was apparently ready to defy the suzerainty of Delhi, presumably because of his getting scent of the internal dissensions that had arisen at the Mughal Court towards the close of Shah Jahan’s rule. For we learn from one of these documents that he took possession of the estate of Jog Chand at the latter’s demise and carried captive some of his relatives, although he knew well that they were under the protection of the Emperor. He does not appear to have surrendered the tract he had seized, notwithstanding Dara Shikoh’s farman issued to him in A. H. 1067 with the threat that “if from obstinacy and imprudence he deferred releasing Jog Chand’s grandson and giving up the district, we would order… Raja Raj Rup… Jahangir Quli Beg and the Faujdar of Jammu that they should go up to the districts of his zamindari and annihilate him”. Raja Raj Rup mentioned here is the Raja of Nurpur, who met Dara at Lahore. Manucci tells us how he was entreated by the unfortunate heir to the Mughal throne. “To gain him more securely to his side, he (Dara) allowed his wife to send for the raja to her harem… she addressed him as her son… and offered him water to drink with which she had washed her breasts, not having milk in them as a confirmation of her words. He drank with the greatest acceptance and swore he would be ever true, and never fail in the duties of a son.” He is stated by Manucci to have obtained ten lakhs from Dara to enlist soldiers. Notwithstanding all this, he proved faithless and was won over by Aurangzeb. He was the son of Raja Jagat Singh of Nurpur and not an uncle of Raja Jaswant Singh of Jodhpur, as is stated in the Sirmur State Gazetteer. The name of Jahangir Quli Beg does not occur in the list of the principal Generals given by Manucci. It may, therefore, be surmised that he was a General of secondary rank.
A perusal of what has been said above shows how history repeats itself. What happened in the Ravi Valley about the tenth century was repeated in the Bias valley in the sixteenth century. Both Chamba and Kulu, before being consolidated into states, were governed by petty Ranas and Thakurs, each supreme in his own sphere. The ruler of the upper valley conquered the lower part and this led to the removal of the seat of government from Brahmor to Chamba in one case and from Jagatsukh to Sultanpur in the other. The earlier rulers worshipped Devi, as is evidenced by Meruvarman’s images of Lakshana Devi and Śakti Devi in the Ravi Valley and by the sanctuaries of Sandhya and Hirma in Kulu. In later days Vishnuism became State religion. This is more clearly marked in Kulu, where Jagat Singh made Raghunath the real Maharaja or ruler of the State, whereas devatas became his vassals and once a year had to attend court at Sultanpur, a practice which continues down to the present time (Plate LXXXV).
Though resembling each other in their history, these sister valleys have been very dissimilar in culture. Chamba has proved to be an inexhaustible mine of inscriptions, some of which rank with first-class Sanskrit compositions, and in this respect has been far superior to Kulu, which has probably never known a period of literary activity. It is in point of conquests that Kulu far surpassed its rival States. In its palmy days, we have seen, it held sway over Lahul, Spiti, Bushahr, Bangahal and Saraj; and even Mandi had to submit to it.
Finally, it will not be out of place to note that, though poor in written records, the Kulu valley is very rich in legends and traditions highly interesting to students of folklore. The sages as well as the heroes of the epics of ancient India have their shrines here, and curious legends are attached to them. Above Jagatsukh a cave is still pointed out where Arjuna passed his days of asceticism, when Śiva appeared to him in the form of a savage Kirata. Hirma, the man-eating rakshasi Hidimba of the Mahabharata, is here worshipped as a goddess, and was once the presiding deity of the valley.
Postscript.— To the documents above discussed may be added certain papers in the Archives of Chamba State which have lately become available and are now being examined by Dr. Hutchison. A list of them will shortly appear in an Appendix to the Chamba State Gazetteer and in the Catalogue of the Bhuri Singh Museum, J. Ph. V.
 Harcourt, “Himalayan Districts of Kooloo, etc.,” pp. 7-8.
 Kangra Gazetteer, Part A (1904), p. 48.
 Ancient Geography, p. 142.
 Cf. The popular sayings Kulu ke ullu and Gaye Kullu hoe ullu.
 Book VI, Canto IX.
 Canto 55, sts. 48-52. The reading uluta is a clerical error. Cf. Rapson J.R.A.S. 1900, p. 531.
 Chap. 14, sts. 29-30. Some of the names in these lists are tribal, e.g. Kira: but it can be presumed that they are intended to designate the regions inhabited by those tribes.
 Rajatarangini II, 435-6.
 P. 204 (Bombay, 1896).
 Jonaraja, Rajatarangini verse 1108. It may be added in passing that Mr. J. C. Dutt’s rendering “Kulutanagarim” into “the city of Luta” is evidently wrong. Cf. J. C. Dutt, Kings of Kasmira Vol.III, App. p. XXII and Duft, Chronology of India, p. 315.
 Si-yu-ki, Beal’s translation, Vol. I, p. 177.
 Ancient Geography, p. 142.
 Harcourt, Op. cit. p. 115, and Kangra Gazetteer, 1897, Parts II to IV, p. 19.
 On the mixture of Lamaism and Hinduism in these regions see Dr. vogel’s article Triloknath in J.A.S.B. Vol. LXX, part I, p. I.
 Captain Harcourt has used the old spelling which makes it difficult to grasp the correct pronunciations of these names. Changes have, therefore, been made to give such names in the right form.
 Dr. Hutchison conjectures that Meru or Meruvarman is meant. He uses the spelling Umer which he explains as a transposition of Meru.
 Kangra Gazetteer, 1897, Part II-IV, pp. 26-27.
 J. A. S. B., Vol. LX, 1891, p. 97 ff.
 A. H. Francke, History of Western Tibet, p. 63 ff.
 Cross between yaks and cows according to Dr. marx. It is strange to find Dzos mentioned as tribute for they are not found in Kulu.
 Mr. A. H. Francke tells me that they were not tax collectors; it was the trade contract which required auch payments.
 Mr. A. V. Bergny, J. R. A. S., 1900, p. 415 and p. 537.
 Annual Progress Report, Punjab and U. prov., 1904-5, p. 14, No. 66. I am indebted to Mr. D. R. Bhandarkar for the decipherment of this epigraph.
 A. S. R., 1902-03, pp. 255 ff.
 Op. cit., p. 257.
 A. S. R., 1903-04, p. 266.
 Mudrarakshasa (ed. Telang), pp. 48 and 407.
 Moorcroft, Travels in the Himalayan Provinces, Vol. I, p. 184.
 It should be noted that Makarsa is nowhere mentioned in the Mahabharata.
 Majma-tawarikh-riyasat-hai-Kohistan-Panjab, Part III, pp. 29-30.
 Cf. Annual Progress Report, Punjab and U. Prov., for 1904-05, pp. 12 ff. Nos. 50-59.
 A copy of one of these Sanads which is kept in our office gives Badshah Ghazi Muhammad ibn Saif(?) Dara Shikoh as the superscription. A comparison with other letters shows that this is wrongly copied. No king Muhammad is known as the son of a Sai Dara Shikoh. Another mistake is also to be noticed in this copy. It gives A. H. 1011 as the date of the sanad which is altogether impossible, for it was Akbar who ruled then.
 Bernier, Voyages (Amsterdam 1699), Vol. I, pp. 84 f. Manucci, Storia do Mogor (transl.-W. Irvine), Vol. I, p. 271. Cf. also Sirmur State Gazetteer (Lahore 1907) p. 13, where a farman of A. H. 1069 is quoted in which Aurangzeb addressed similar instructions to Raja Subhag Parkash of Sirmur.
 Op. cit., Vol. I, p. 310.
 Cf. A. S. R., 1904-05, p. 114.
 Sirmur State Gazetteer, 1904, p. 13.
 The kardar or manager of the Sita Ram temple at Gojra, kothi Jagatsukh, also possesses some sanads. They are said to be dated between Śastra Samvat 95 and 5, i.e., between A.D. 1719 and 1729, but I have not been able to inspect them.