Pramāṇa · Search · About · API

The Dynastic History of Northern India (Early Mediaeval Period) v2.c28: Chapter XI. Candrātreyas (Candellas) of Jejā-Bhukti (Bundelkhand)

The Dynastic History of Northern India (Early Mediaeval Period) v2.c28

Language
English

Contents

Chapter XI. Candrātreyas (Candellas) of Jejā-Bhukti (Bundelkhand)

v2.c28.2

There is little doubt that the rise of the Candella power in the south of the Jumna was one of the important contributory causes that hastened the downfall of the Gurjara-Pratīhāra empire in Northern India. When however we try to trace the first beginnings and growth of the Candella kingdom we meet with considerable difficulties. Tradition records the following account of their rise.

v2.c28.3

Hem-rāj, the *purohit* of Indrajit, the Gaharwar *rājā* of Benares, had a daughter named Hemāvatī. She "was very beautiful, and one day when she went to bathe in the Rāti Tālāb, she was seen and embraced by *Candramā* (the Moon); as he was preparing to return to the skies, Hemāvatī cursed him. 'Why do you curse me?' said Candramā, 'your son will be lord of the earth, and from him will spring a thousand branches.' Hemāvatī enquired: 'How shall my dishonour be effaced when I am without a husband?' 'Fear not,' replied Candramā, 'your son will be born on the bank of the Karṇavatī river; then take him to Khajurāya, and offer him as a gift and perform a sacrifice. In Mahoba (Mahotsavanagara) he will reign, and will become a great king . . . . . On the hill of Kālañjara he will build a fort'..."¹ A copy of *the Mahobā-Khaṇḍ* consulted by Cunningham, gave the date of the coronation of this child, who was named Candra-varmā, as *Saṁvat* 225.¹ The traditions preserved by the family of the Qānungo of Mahoba inform us that Candra-varmā overthrew the Parihārs in Bundelkhand.² The genealogies preserved in the official charters of the Candella princes, however, are ignorant of the existence of this prince. Thus the Khajraho stone inscription of Dhaṅga (V.S. 1011) gives the following account of the origin of the dynasty³:—

v2.c28.4

¹ *ASR*, Vol. II, pp. 445-46.

v2.c28.5

¹ *Ibid.*

v2.c28.6

² *JASB*, 1881, Vol. L, p. 3. The date of this event is given as *Saṁvat* 677.

v2.c28.7

³ *EI*, Vol. I, pp. 122 ff. This is the earliest record of the line, as far known, which supplies any information about the origin of the family. There are slight variations in other records. Thus in the Khajraho stone inscription of Kokkala (V. S. 1058) we have: Atri: From his eye the Moon; His son the sage Candrātreya:—In the family of the *Candrātreya-vaṁśajaḥ* princes, Nannuka; see *ibid*, pp. 137 ff. The copper plates usually begin with the praise of *Candrātreya-narendrāṇāṁ vaṁśa* and then give the name of the grandfather of the donor of the grant; see *IA*, XVI, pp. 207 ff. Sometimes the inscription begins with the praise of only the Moon, the progenitor of the *Candrātreya-narendrāṇāṁ vaṁśa*; see *EI*, Vol. XVI, pp. 9 ff. A fragmentary Mahoba inscription, however, begins with the genealogy with "From the beloved of the night there sprang a race beloved by all." The preserved portion does not contain the word Candra or Candrātreya; *ibid*, Vol. I, pp. 217 ff. Generally speaking, the earliest inscriptions refer to the dynasty as *Candrātreya-vaṁśa*. But the Dudahi stone inscriptions of Devalabdhi, a grandson of Yaśovarman, describe the family as *Candrell-ānvaya*, from which no doubt the later form Candella is derived. Kielhorn suspected that the name *Candrātreya* was really a later Sanskritised form of the word *Candrella*. According to him this last word was formed from Candra by the addition of the Prakrit suffix *illa*. The form *Caṁdella* occurs in the Deogarh rock-inscription of Kīrtivarman, *IA*, Vol. XVIII, pp. 237 ff. The form Candella is found in the Benares grant of the Kalacuri Lakṣmī-Karṇa *EI*, Vol. II, p. 306; while the form *Candela* occurs in the Madanpur inscription of Cāhamāna Pṛthvīrāja III, (V.) *Saṁvat* 1239; see *ASR*, Vol. XXI, p. 174. We reserve for discussion the question of the origin of the Candellas in Vol. III of the present work.

v2.c28.8

``` Viśvasṛk Purāṇapuruṣa (The creator of the universe, that

v2.c28.9

| ancient being).

v2.c28.10

Sages Mārīci, Atri, etc.

v2.c28.11

Muni Candrātreya...acquired fierce might by ceaseless

v2.c28.12

| austerities

v2.c28.13

From him princes (bhūbhujām) who had the power to destroy or protect the whole earth.

v2.c28.14

In this family Nṛpa Nannuka

v2.c28.15

Vākpati ______| . . . . .

v2.c28.16

Jayaśakti Vijayaśakti

v2.c28.17

Rāhila

v2.c28.18

Harṣa = Kañcukā

v2.c28.19

Yaśovarman

v2.c28.20

Dhaṅga (V.S. 1011 = A.D. 954) ```

v2.c28.21

It should however be observed that both tradition and inscriptions agree in tracing the descent of the family to the Moon. The fact that the traditional founder of the dynasty is first taken to Khajraho¹ is also consistent with epigraphic and foreign evidence. For the earliest inscriptions of the Candellas come from that place, and the *Kāmil*, one of the oldest compilations of Arab history, connects the dynasty with Kajurāhah (كجوراهه).² Cunningham discovered another agreement between tradition and epigraphic evidence. By counting backwards from 954 A.D., the earliest known date of Dhaṅga, the 6th lineal descendant of Nannuka, and assigning a period of 20 to 25 years for each generation, he fixed upon the first quarter of the 9th century A.D., as the approximate date of Nannuka.³ Then by referring *Saṁvat* 225, one of the traditional dates for the foundation of the Candella dynasty, to the Harṣa era, he found it in perfect agreement with the above date (225+606=831 A.D.).⁴ There is yet another agreement between tradition and epigraphy if we accept the name Candra-varmā as a mere *biruda* of Nannuka. But the statement that the founder of the dynasty supplanted the Parihārs cannot be accepted. From about the time of Nannuka down to the beginning of the 10th century the Gurjara-Pratīhāras were at the height of their power, and it is unlikely that, during this period, the Candellas would succeed in driving them out of the land lying to the south of the Jumna. It is more likely that the Candella Nannuka was at first a feudatory of Nāgabhaṭa II (*c.* 815-33 A.D.) and ruled over a small principality round about Khajraho, the Kharjjuravāhaka of the inscriptions. This agrees with the tradition which places the original home of the Candellas at Maniyagarh in the Chhatarpur State (C. I.). In the Khajraho inscription of Dhaṅga, Nannuka is called *nṛpa* and is described as ' a touchstone to test the worth of the gold of the regal order, who playfully decorated the faces of the women of the quarters with the sandal of his fame,' and whose enemies carried ' his commands on their heads, like a garland.'¹ In another Khajraho inscription of the same king Nannuka is called a *mahīpati* 'whose prowess reminded the gods of Arjuna.'² This is the only information available about the first historical personage in the genealogy of the Candellas. The Indian *praśastikāras* at any rate do not err on the side of moderation and it is extremely impossible that they would have omitted such a tangible historical fact as the defeat of the Parihāras, if that incident really happened in the career of the founder of the dynasty, and attached to his memory such vague generalities which in Indian records mean practically nothing. It is also significant that Nannuka's name is omitted from all the other Candella inscriptions, which are not few. He would have certainly loomed as a more substantial figure in the family tradition if he had really laid the foundation of the sovereignty of the family by violently uprooting another dynasty. The evidence available therefore suggests that his position was that of a petty feudatory ruler. It would even be risky to assume that he 'enjoyed some share of sovereign power' from the mere fact that in the inscriptions of his distant descendants he is designated a *mahīpati* or a *nṛpa*. There is also no evidence to prove the alleged connection of the founder of the dynasty and the *purohita* of the Gāhaḍavālas. In fact epigraphic evidence is against the supposition that Nannuka could even be a contemporary of the early Gāhaḍavālas of Benares (11th century A. D.).¹

v2.c28.22

¹ Sometimes spelt Khajuraho.

v2.c28.23

² *TKA*, Bulak, 1874, Vol. IX, pp. 115-16.

v2.c28.24

³ *ASR*, Vol. II, p. 447.

v2.c28.25

⁴ The other traditional dates for the foundation of the Candella power are 204, 661, and 682. We can of course refer the first of these to the Harṣa era; but the three other dates, 661, 677, and 682 cannot be satisfactorily explained. See *JASB*, Vol. L, pp. 3-6.

v2.c28.26

¹ *EI*, Vol. I, p. 125, V. 10.

v2.c28.27

² *Ibid.*, p. 141, Vs. 14-15.

v2.c28.28

¹ See *ante*, DHNI, Vol. I, pp. 504 ff.

v2.c28.29.1

The next person in the Candella genealogy is Vākpati, the son of Nannuka. His existence is also known from the two inscriptions referred to above. The first informs us that the illustrious Vākpati defeated his enemies in battle and made the Vindhyas his pleasure-mount (*Krīḍā-giri*).² According to the other record, the Kṣitipa Vākpati is said to have excelled by his wisdom and valour (even) the mythical kings Pṛthu and Kakustha.³ From the fact that he is referred to as having made the Vindhyas his pleasure-mount we may perhaps be justified in concluding that Vākpati succeeded in extending to some degree the limits of his small ancestral principality. But in my opinion the persons who really succeeded in establishing the foundation of the prosperity, though not the sovereignty of the family, were his sons Jayaśakti and Vijayaśakti.⁴ Excepting in the two inscriptions mentioned above, these two brothers are generally described in records as the first ancestors of the Candellas. Verse 10 of a fragmentary inscription at Mahoba again tells us that Jejā (Jayaśakti) gave his name to *Jejā-bhukti*, just as Pṛthu did to *Pṛthivī*.⁵ This Jejā-bhukti⁶ is generally accepted as the 'old name of Bundelkhand and the original of the vernacular form Jajāhūtī or Jajāhotī, just as modern Tirhut is derived from Tīra-bhukti.’¹ Cunningham, who first proposed this identification, further identified this name under its vernacular form with the *Chih-chi-t'o* of Yuan Chwang.

v2.c28.29.2

But as the date of Jayaśakti cannot be earlier than about the beginning of the 9th century, and as there is nothing to discredit the epigraphic evidence that he gave his name to the *Bhukti*, we must give up this identification.² Watters may be right when he suggests that the *Chih-chi-t'o* of Yuan Chwang represents modern Chitor and not Jajhoti.³ From the fact that his dominion is called only a *Bhukti* we may infer that Jayaśakti was at best a mere feudatory perhaps of the great Gurjara emperor Bhoja (*c.* 836-88 A.D.). In the inscriptions he is generally praised vaguely for victories over his enemies. Thus in a Khajraho inscription of Dhaṅga we are told that “by the unmeasured prowess” of him and his younger brother “adversaries were destroyed, as woods are burnt by a blazing fire.”⁴ But in another Khajraho inscription of the same king, the younger brother Vijaya is said to have, ‘like Rāma, on his warlike expeditions reached even the southernmost point of India.’⁵ It is not impossible that there may be some historical truth hidden in this reference. But if he really invaded peninsular India he must have done so as the feudatory of some more powerful sovereign.

v2.c28.29.3

For though the brothers are repeatedly referred to in the inscriptions as having added ‘radiance’ to the family of the Candrātreya princes, it is significant that they are seldom given the epithets of sovereign rulers, and are on the contrary often simply called heroes (*vīra*).¹ We may therefore conclude with some probability that Vijayaśakti, who succeeded his elder brother as the head of the family, was still a subordinate chief, possibly of the Gurjara-Pratīhāra Bhoja or his son Mahendrapāla (*c.* 893-907 A.D.). The position of Rāhila, Vijayaśakti's son and successor, does not appear to have been different from that of his father. He is mentioned only in two of the Candella inscriptions so far discovered; and in both of them he is merely praised in conventional phrases for his bravery and victories. Thus in the Khajraho inscription of Dhaṅga, the royal *praśasti-kāra* says: "Thinking of whom (Rāhila) the enemies enjoy little sleep at night.

v2.c28.29.4

Who never tired, at the sacrifice of battle, where the terribly wielded sword was the ladle, where the oblation of clarified butter was made with streaming blood, where the twanging of the bow-string was the exclamation of *vaṣaṭ*, (and) at which exasperated warriors marching in order were the priests, successful with his counsels (as with sacred hymns) sacrificed, like beasts, the adversaries in the fire of enmity, made to blaze up high by the wind of his unappeased anger."² In another Khajraho epigraph of the same king Rāhila is given credit for favouring his friends and punishing his enemies.³ It was not till the reigns of his son Harṣa and his grandson Yaśovarman that circumstances favoured the growth of a strong Candella state. The continuous struggle between the Gurjaras and the Rāṣṭrakūṭas which had been raging since the days of Vatsa and Dhruva suddenly took a bad turn for the Gurjara-Pratīhāras sometime between 915-18 A.D.⁴ The Rāṣṭrakūṭas under Indra III captured Ujjayinī and overran Northern India. The imperial city of Kanauj itself was completely devastated. There is some evidence to show that the Deccani kings held some portion of the Gaṅgā-Yamunā *Doāb* even as late as 930 A.D. This severe blow and the strangle-hold on the very heart of the imperial power must have greatly weakened the control of the central government on its provinces. It is true that the Pratīhāra emperor was assisted by a number of his feudatories to recover some of his dominions.

v2.c28.29.5

A fragmentary Candella inscription at Khajraho, for instance, says that either Harṣa or his son Yaśovarman placed Kṣitipāladeva again on the throne¹ (*Punar-yena-śrī-Kṣitipāladeva-nṛpatiḥ siṁhāsane sthā*[*pitaḥ*]). In their official records, also, till the year 954 A.D., in the reign of Dhaṅga, the Candellas actually acknowledged the sovereignty of the Gurjara-Pratīhāras. But there is enough evidence to show that their obedience to the Kanauj power was fast growing nominal. Harṣa, who was possibly reigning when the Rāṣṭrakūta invasion took place, does not appear to have indulged in any open acts of hostility to the imperial power. In a fragmentary stone inscription discovered at Khajraho, he is said to have "conquered many proud enemies" (*dṛptārivṛndaḥ*).² The Khajraho stone inscription of Dhaṅga dated in V.S. 1011 tells us that "that most excellent of rulers was afraid to offend against the law (*dharma*) and anxious to worship the feet of (Viṣṇu), the enemy of Madhu....(He) married a suitable lady of equal caste (*savarṇā*), named Kañcukā, sprung from the Cahamāna family."³ In the Nanyaura plate of the same king, (V.S. 1055) Harṣa is said to have inspired terror by his terrific array of troops, and made tributary to himself other sovereigns.⁴ Dhaṅga's Khajraho inscription of (V.)S.

v2.c28.29.6

1059 tells us that Harṣa protected the whole earth after subduing his adversaries.¹ These references to Harṣa do not contain any historical facts, but in spite of their vague character they seem to indicate that his was a more substantial figure than any of his predecessors. If Kielhorn was right in his assumption that ‘ the relative *yena* in line 10 ’ of the fragmentary Khajraho inscription mentioned above ‘ refers to Harṣadeva,’ the latter's claim to have reinstalled Kṣitipāla, his overlord, on his throne must indicate a considerable degree of power and prestige. He further strengthened his position by forming a matrimonial alliance with the Cāhamānas, who were also gradually breaking away from the empire. Another matrimonial alliance of Harṣa seems to be revealed by the Benares grant of Lakṣmī-Karṇa, the Kalacuri king of Tripurī. If the *Citrakūṭa-bhūpāla* of this inscription to whom Kokkalla claims to have granted ‘ freedom from fear ’ is identified with the Candella Harṣa, then it looks likely that the Candella princess Naṭṭā (*Naṭṭākhyadevī*), the queen of Kokkalla, was a relative of this Candella prince.³ Without openly defying his overlord, Harṣa thus appears to have slowly consolidated his position at home and abroad. It was left to his son Yaśovarman, *alias* Lakṣmavarman, to engage in more ambitious projects for increasing the Candella territory by launching into schemes of foreign conquest and aggrandisement. I have already shown⁴ that some time before 940 A.D.

v2.c28.29.7

the Rāṣṭrakūṭas' attacks on the Gurjara-Pratīhāra empire were renewed. The Jura inscription of Kṛṣṇa III, discovered recently in the Maihar state, shows that his claim to have captured Kālañjara sometime before 940 A.D., from the Gurjaras, may not be entirely baseless. That the Gurjaras referred to in the Karhad and Deoli plates in connection with Kālañjara, were the Gurjara-Pratīharas of Kanauj is proved by the Barah copper plate of Bhojadeva, which granted land in *Kālañjara-maṇḍala* of the *Kanyakubja-bhukti*.¹ The Khajraho stone inscription of Dhaṅga, dated in 1011 V.S. (A.D. 954) tells us that Yaśovarman easily conquered the mountain of Kālañjara, ‘the dwelling-place of Śiva.’² As Yaśovarman therefore appears to have been a contemporary of Kṛṣṇa III, it is not unlikely that he may have captured the famous hill-fort, not from the Gurjara-Pratīhāras, as was so long believed, but from the Rāṣṭrakūṭas. The sudden emergence of the Candellas as one of the first-class powers in the country south of the Jumna may therefore be due partially to their success against these ruthless marauders of the Deccan, the predecessors of the Marāthā horsemen of a later period. In the Khajraho inscription, referred to above, the poet gives the following graphic, if somewhat exaggerated, account of the victories and campaigns of Yaśovarman:

v2.c28.30

² *EI*, p. 125, Vs. 11-12.

v2.c28.31

³ *Ibid*, p. 141, Vs. 16-17.

v2.c28.32

⁴ There are variant forms of these two names in the inscriptions. For the former we have: Jejā (*EI*, Vol. I, p. 221, V. 10) and Jejjāka (*EI*, Vol. I, p. 122, line 6); for the latter, we have Vijaya (*EI*, Vol. I, p. 141, V. 18), Vijjā (*EI*, Vol. I, p. 122, line 6) and Vijā (*EI*, Vol. I, p. 221, V. 10).

v2.c28.33

⁵ *EI*, Vol. I, p. 221.

v2.c28.34

⁶ There are variants of this name. A Madanpur stone inscription gives the form *Jejāka-bhukti* (Cunningham, *ASR*, Vol. X, plate XXXII, No. 10, lines 2-3). The same form of the name is found in several other Madanpur records, with the addition of *deśa*, or *maṇḍala*. (See *ibid*, Vol. XXI, p. 174.) The Ratnapur stone inscription of the Kalacuri Jājalladeva, (Cedi) Saṁvat, 866, gives the form *Jejābhuktika*; see *EI*, Vol. I, p. 35, V. 21.

v2.c28.35

¹ *EI*, Vol. I, p. 218; *AGI*, 2nd ed., Calcutta, 1924, p. 551. The form *Jejāhūti* with its capitai Kajurāha occurs in Birūnī's *Indica*, Trans. by Sachau, Trübner, Vol. I, p. 202. Ibn Batuta visited Kajarrā in A. D. 1335. This city must be the same as the Kajurāhah of Bīrūnī and Ibn ul-Athir. Cunningham in his footnote, on p. 551, gives the name of the city visited by Ibn Batuta in the ‘original Persian (a slip for Arabic) characters' as Kajurā. But in the text edited by Defrémery and Sanguinetti (*Voyages d'ibn Batoutah*, Paris 1858, p. 39) it is clearly written as كجرا ; see also *Ibn Batuta*, by Gibb, Routledge, 1229, pp. 226 and 363.

v2.c28.36

² Untenability of this identification of Cunningham was first pointed out to me by Prof. R. C. Majumdar of the University of Dacca.

v2.c28.37

³ *YC*, Vol. II, p. 251.

v2.c28.38

⁴ *EI*, Vol. I, p. 126, V. 15.

v2.c28.39

⁵ *EI*, Vol. I, pp. 141-42, V. 20.

v2.c28.40

¹ *Ibid*, IV, p. 157, line 2; Vol. X, p. 47, line 2; Vol. XVI, p. 12, line 3, and p. 274, line 2.

v2.c28.41

² *EI*, Vol. I, p. 181, V. 17, also p. 126. This idea of comparing battles to sacrifices is probably taken from the *Mahābhārata*; see *Baṅgavāsī* (Calcutta) ed., *Udyogaparva*, Chap. 141, Vs. 29-51; see also *JL*, Vol. XIV, 1927, *Notes on War in Ancient India*, p. 1 ff.

v2.c28.42

³ *EI*, Vol. I, p. 242, Vs. 22-28.

v2.c28.43

⁴ See *DHNI*, Vol. I, pp. 579 ff.

v2.c28.44

¹ *EI*, Vol. I, p. 122, line 10; see also *DHNI*, Vol. I, pp. 581-82.

v2.c28.45

² *EI*, Vol. I, p. 121, line 7.

v2.c28.46

³ *Ibid*, Vol. I, p. 126, Vv. 20-21.

v2.c28.47

⁴ *IA*, Vol. XVI, pp. 202-06, lines 2-3.

v2.c28.48

¹ *EI*, Vol. I, p. 142, Vs. 25-28. This inscription also mentions his wife Kañcūkā, but it does not refer to her lineage; see *ibid*. pp. 142-43, V. 29.

v2.c28.49

² Kielhorn suggested that Harṣa 1st first defeated Kṣitipāladeva and subsequently reinstalled him on his throne. But in view of recent discoveries and the identification of Mahīpāla and Kṣitipāla, it would be better to accept the view that the vanquisher of the Kanauj monarch was the Rāṣṭrakūṭa Indra III and not the Candella chieftain; see *EI*, Vol. I. p. 122.

v2.c28.50

³ *EI*, Vol. II, p. 306, V. 7; also p. 301. The acceptance of the suggested identification would show that this *Citrakūṭa* must be differentiated from the hill of the same name in Rajputana and identified with the hill bearing the same name in Bundelkhand. Cunningham identified a hill ‘situated on the left, or west, bank of the Paisuni river, about 5 miles to the S.W. of Karwi, and 50 miles to the S.E. of Banda,’ with the Citrakūṭa of the *Rāmāyaṇa*; *ASR*, Vol. XXI, pp. 10-13. See also *DHNI*, *infra*, chapter on the Haihayas.

v2.c28.51

⁴ See *DHNI*, Vol. I, pp. 588 ff.

v2.c28.52

¹ *EI*, Vol. IV, p. 284; *JBRAS*, Vol. XVIII, p. 247; *EI*, Vol. XIX, p. 18, line 6.

v2.c28.53

² *EI*, Vol. I, pp. 122 ff., V. 31.

v2.c28.54

“Who was a sword to (cut down) the Gauḍas as if they were pleasure-creepers (*krīḍālatā*), equalled the forces (*bala*) of the Khaśas (and)³ carried off the treasure of the Kośalas, before whom perished the Kāśmīrī warriors (*naśyat-Kāśmīra-vīraḥ*); who weakened the Mithilas (*śithilita-Mithilaḥ*), (and) was as it were a god of death to the Mālavas (*Kālavan Mālavānām*), who

v2.c28.55

³ The settlement of the Khaśas during this period extended in a wide semicircle from Kaṣṭavār in the south-east to the Vitastā valley in the west. The hill states of Rājapurī and Lohara were held by Khaśa families. Stein identified the Khaśas with the ‘modern Khakha tribe, to which most of the petty hill chiefs and gentry in the Vitastā valley below Kaśmīr belong.’ See Stein's Eng. Trans. of the *Rājataraṅgiṇī*, Vol. I, pp. 47-48, footnote on I, 317, and Vol. II, p. 430.

v2.c28.56

brought distress on the shameful Cedis (*Sīdat-sāvadya-Cediḥ*), who was to the Kurus what a storm is to the trees (*Kuru-taruṣu marut*), and a scorching fire to the Gurjaras (*Saṁjvaro Gurjarāṇām*)" (V. 23).

v2.c28.57

"Free from fear, he impetuously defeated in battle the Cedi king whose forces were countless......" (V. 28).

v2.c28.58

"At the conquest of the regions, his soldiers gradually managed to ascend the slopes of the snowy mountains, where plentiful flowers had been gathered by Umā from every tree of paradise, where the troops of horses became unmanageable with fright at the gurgling sound of the torrents of the Ganges......" (V. 30).

v2.c28.59

"He easily conquered the Kālañjara mountain, the dwelling-place of Śiva, which is so high that it impeded the progress of the sun at midday" (V. 31).

v2.c28.60

"The illustrious Lakṣavarman in his conquests of the regions made, equal to Indra, the daughter of Kalinda (Yamunā) and the offspring of Jahnu (Ganges) one after another his pleasure-lakes, encamping the forces of his army on either banks unmolested by any adversaries (and rendering) their waters muddy with the bathing of his furious mighty elephants" (V. 39).¹

v2.c28.61

¹ *EI*, Vol. I, pp. 126-28 and 132-34.

v2.c28.62

According to these verses, therefore, Yaśovarman dominated the whole region from the Himalayas to Malwa and from Kashmir to Bengal. As I have already pointed out, much of this is no doubt mere *praśasti*, and poetic exaggeration. Thus the expression *Kośalaḥ Kośalānām* seems rather to convey a desire on the poet's part to show his skill in punning than an historical sense. But there is no inherent impossibility in the assumption, that he really raided a large portion of northern India. His victory over the Cedi king, which is mentioned twice, looks like a genuine fact; and it is likely that one of the earlier Kalacuri princes of Tripurī, possibly Lakṣmaṇarāja or his predecessor Yuvarāja I, was the *Cedirāja* whose 'countless forces,' were defeated by Yaśovarman after a fierce contest. The contemporary Gauḍa king was probably either Rājyapāla or his son Gopāla II. In Kashmir reigned at this period a series of comparatively unimportant princes beginning with Cakravarman (923-37 A.D.), and ending in Parvagupta (949-50 A.D.) while the Khaśa state of Lohara was ruled by one of the predecessors of Siṁharāja (*c.* 950 A.D.), possibly Candurāja. The territories of the Mālavas, Kośalas, and Kurus appear to have been still under the Gurjara rulers of Kanauj, while Mithilā at this period may have been recovered by the Pālas of Bengal and Bihar.¹ If there is any basis of fact in these claims of successful invasions and victories it is clear that he soon became a standing menace to the imperial house at Kanauj. The expression *Saṁjvaro Gurjarāṇāṁ* suggests that he may have even come into violent conflict with his former overlords. But the Khajraho inscription of Dhaṅga dated in V.S. 1011 shows that though practically independent, the Candellas still retained in their official documents the fiction of their subservience to the Gurjara-Pratīhāras.²

v2.c28.63

¹ In the time of Mahendrapāla, the Gurjara-Pratīhāra dominions extended from Śrāvastī in U.P. right up to Paharpur in N. Bengal. See *DHNI*, Vol. I, chapter on the Dynasties of Bengal and Bihar, pp. 569-70. The Asiatic Society’s plate of Vināyakapāla, which was issued from Mahodaya. and grants land in Vārāṇasī-*Viṣaya*, in the Pratiṣṭhāna-bhukti, in V.S. 988, the Rakhetra stone-inscription of the same prince discovered in the Gwalior region dated in V.S. 999-1000, and the Partabgarh stone-inscription of Mahendrapāla (II) dated in V.S. 1003 seem to indicate that the Gurjara-Pratīhāras during the period *c.* 931-46, held a large portion of the irregular quadrilateral formed by Mandu, Gwalior, Kanauj and Benares.

v2.c28.64

² See *DHNI*, Vol. I, p. 589.

v2.c28.65

The Khajraho inscription referred to above informs us that Yaśovarman erected a ‘charming splendid home of (Viṣṇu), the enemy of the Daityas, which rivals the peaks of the mountain of snow.’ We are told that the image of Vaikuṇṭha which he set up in this temple was first obtained by ‘the lord of Bhoṭa (*Bhoṭanātha*) from the Kailāsa, and from him Sāhi, the king of Kīra received it as a token of friendship; from him afterwards Herambapāla obtained it for a force of elephants and horses, and (Yaśovarman himself) received it from the *Hayapati* Devapāla, the son of (Herambapāla)¹ The temple mentioned above has been identified by Cunningham with the Vaiṣṇava temple at Khajraho variously known as that of Rāmacandra, Lakṣmaṇa or Caturbhuja.² The credit for building this temple, together with that for excavating a big tank (*taḍāgārṇavam*), is given to Yaśovarman by verses 38 and 39 of another Khajraho inscription of Dhaṅga dated in V.S. 1059.³

v2.c28.66

¹ *EI*, Vol. I, pp. 129 and 134, Vs. 42-43. For the identification of Herambapāla and Devapāla, see *DHNI*, Vol. I, chapter on the *Later Gurjara-Pratīhāras*, pp. 571 ff. Bhoṭa during this period indicated Tibet while the *Kīras* must be located somewhere near Kashmir. *The Bṛhatsaṁhitā* (XIV. 29) places them with the Kaśmīras in the N.E.

v2.c28.67

² *ASR*, Vol. II, pp. 425-27. The temple is 85 ft. 4 inches in length and 44 ft. in breadth. The stone-inscription of Dhaṅga which refers to the building of this temple was originally discovered at the base of the Caturbhuja temple, and is now built into the wall inside its entrance porch. *EI*, Vol. I, p. 144.

v2.c28.68

³ *Ibid*, Vol. I, p. 144.

v2.c28.69

Yaśovarman had a son named Dhaṅga by his queen Puppādevī.⁴ The Dudahi stone inscriptions reveal the existence of probably another son of Yaśovarman. These records, six in number, were discovered in a temple of the village of Dudahi in the Lalitpur district of U. P. (Long. 78°27′E., Lat. 24°24′N.). We know from these inscriptions that the temple was erected by the illustrious Devalabdhi belonging to the *Candrellānvaya*, the son of the illustrious Kṛṣṇapa and Āsarvā and the grandson (*naptṛ*) of *Mahārājādhirāja* Yaśovarman.⁵ Cunningham's identification of this Yaśovarman with Candella Yaśovarman of Khajraho has been generally accepted by scholars.⁶

v2.c28.70

⁴ *Ibid*, V. 40. Is this name the Prākṛt form of *Puṣpa-devī* ?

v2.c28.71

⁵ First edited by Cunningham with the photo-zincographs of five of them in *ASR*, Vol. X, pp. 94-95, and plate XXXII, Nos. 1, 2 and 4-6. They were then re-edited by Kielhorn in 1989 in the *IA*, Vol. XVIII, pp. 236-237.

v2.c28.72

⁶ The undated fragmentary Khajraho stone-inscription discovered near the temple of Vāmana may belong to Yaśovarman or his father Harṣa. See *EI*, Vol. I, pp. 121-22; *JRAS*, 1904, p. 654, fn. 1.

v2.c28.73

Yaśovarman was succeeded by Dhaṅga sometimes before V.S. 1011, the earliest known date of the latter's reign. The inscription which carries this date attempts to give us an idea of the extent of Dhaṅga's dominions. We are told that he "playfully acquired by the action of his long and strong arms, as far as Kālañjara and as far as Bhāsvat, situated (?) on the banks of the river of Mālava (*Mālava-nadī*); from here also to the bank of the river Kālindī, and from here also to the frontiers of the Cedi country (*deśa*) and even as far as that mountain called Gopa (*Gopābhidhāna-giri*).'"¹ As contrasted with the vague claims of victories from Kashmir to Bengal and the Himalayas to Malwa in the case of his father, this comparatively modest specification of the boundaries has the appearance of being based on fact. Though his *praśastikāra* tells us that this wide dominion was acquired by Dhaṅga's strong arms alone, he was no doubt greatly aided by the campaigns of his father. The same inscription which mentions these boundaries of his dominions contains in the last line the following statement:—

v2.c28.74

¹ *EI*, Vol. I, p. 129, V. 45. Kielhorn identified *Mālavanadi* with the river *Vetravatī* or Betwa and Bhāsvat with Bhailasvamin, the mod. Bhilsa. Gopagri is certainly Gopādri, identified with mod. Gwalior. Unfortunately no inscriptions of the Candellas have yet been discovered in Gwalior but it is likely that the family of the Kacchapaghāta *Mahārājādhirāja* Vajradāman, for whose reign we have the Gwalior image inscription, dated V.S. 1034 (*JASB*, Vol. XXXI, p. 393), may have been feudatories of the Candellas. The Dubkhund inscription (*EI*, Vol. II, p. 237) dated in V.S. 1145 indicates that the Kacchapaghāta Arjuna was a feudatory of Vidyādhara, the grandson of Dhaṅga.

v2.c28.75.1

" 'While the illustrious Vināyakapāladeva is protecting the earth, the earth is not taken possession of by the enemies, who have been annihilated.'"² This statement undoubtedly shows that up to 954-55 A. D., the Candellas retained in their public documents a formal, if nominal, recognition of the imperial line at Kanauj. But it is significant that the names of the Gurjara-Pratīhāra rulers do not occur again in any Candella document after this date, indicating that after 954-55 Dhaṅga became a sovereign ruler *de facto* and *de jure*. The expression 'as far as Kālañjara' which occurs in the quotation cited above shows that up to the date of the inscription the capital of the Candellas was situated not in that hill-fort but in some other city. That this city was Khajraho is probably indicated by an epigraph of Dhaṅga which is dated "in (V.)S. 1059 at Śrī-Kharjjuravāhaka in the realm of Dhaṅga".¹ This is further supported by the *Kāmil* of Ibn ul-Athīr, which refers to Dhaṅga's grandson as ruler of Kajurāha.² It is also significant that the earliest inscriptions of the Candellas including those of Dhaṅga, came from Khajraho and not from Kālañjara³ or Mahoba. It should be noted however that as early as (V.) S. 1055 Dhaṅga is given the epithet *Kālañjarādhipati*.⁴ This may indicate that Kālañjara was regarded as a sort of second capital of the kingdom.

v2.c28.75.2

But it is more probable that the epithet simply indicates the possession of one of the strongest hill-forts of India,⁵ which in an age ignorant of artillery, made his position almost impregnable. The strategic importance of this hill was well-known in India at this time, and the Kalacuris, the Gurjara-Pratīhāras, the Rāṣṭrakūṭas and the Candellas tried in succession to retain their hold on its fortifications. The possession of this fortress, as well as that of Gwalior must have consolidated the position of the Candellas in Central India and encouraged them to attempt the systematic conquest of portions of the Ganges valley. Yaśovarman is already said to have encamped his troops on the banks of the Ganges. The Nanyaura plate of Dhaṅga which was issued from Kāśikā (Benares), shows that this was no idle boast, and that in V.S. 1055 the Candellas were certainly on the Ganges.⁶ The possession of Prayāga (Allahabad) by Dhaṅga is probably indicated by the fact that he is reported to have 'entered into beatitude' by abandoning his body in the waters of the Jāhnavī, and the *Kālindī*.¹ It is not unlikely that from these bases Dhaṅga may have penetrated further eastward in the Ganges valley.

v2.c28.75.3

A Khajraho epigraph tells us that the wives of ‘the kings of Kāñcī, Andhra, Rāḍhā, and Aṅga lingered in his prisons.’² Another verse of the same inscription says that ‘he was so powerful that the rulers of Kośala, Kratha,³ Siṁhala and Kuntala humbly listened to his commands.’⁴ It is therefore possible that he may have even successfully raided some portions of peninsular India. In any case there is enough evidence to show that Dhaṅga had become one of the most powerful rulers of Central India during this period. One of the Candella inscriptions goes so far as to say that he, having defeated on the battlefield the *Kānyakubja-narendra*, obtained exalted sovereignty (or empire, *Sāmrājya*).⁵ If true, this statement shows that the hegemony which the rulers of Kanauj so long enjoyed in Northern India was at last taken away from them by the Candellas during the reign of Dhaṅga. But as ‘Utbī in 409 A.H. (1018 A.D.) describes the Kanauj ruler Rājyapāla as ‘the chief of all princes of India,’⁶ we must conclude that the halo of imperialism still lingered over the brow of the Gurjara-Pratīhāras for at least sometime after Dhaṅga’s death. The accepted interpretation of another statement about Dhaṅga also appears to me to be doubtful.

v2.c28.75.4

A fragmentary Candella inscription discovered at Mahoba thus describes his achievements: “……Dhaṅga, who caused the destruction of his enemies, and who by the strength of his arms equalled even the powerful Hamvīra, who had proved a heavy burden for the earth......”¹ The word *Haṃvīra* is apparently a corruption of the Arabic word Amīr (امير, Commander, leader), derived from the root Amr (امر, command). The word *Amīr* in this sense is found in Islamic history in the title *Amīr ul-mu'minīn*, *i.e.*, ‘lord of the faithful’ which became a regular title of the Caliphs since the days of ‘Umar.² The term gradually came to be applied to kings, and later on to any high official of the state. The coins of the Yamīnīs of Ghazni show that they were known by the title of *Amīr*.³ Stein has shown that the *Hammīra* of Kalhaṇa, who was a contemporary of Sāhi Trilocanapāla and the Kashmirian king Saṃgrāmarāja (1003-28 A.D.) must be identified with Mahmud of Ghazni.

v2.c28.75.5

Thus *Hammīra* appears to be another Indian corruption of the Arabic word *Amīr*.⁴ That this is the correct interpretation is proved by the Sanskrit legend on the reverse of some of the gold coins of Muhammad ibn Sām, which runs as follows:—*Śrīmad Hamīra Mahamad Sām*.⁵ From this time onwards up to the reign of Balban the epithet *Śrī-Hamīra* (*Śrī-Hamīra* or *Śrī-Hammīra*) occurs as the regular designation of the Ghūrī and Delhi Sulṭāns in their coins and inscriptions.⁶ The epithet also occurs on the coins of some of their provincial satraps.⁷ There is no doubt therefore that *Hammīra* *Hamīra*, or *Haṃvīra* was to the Indians the accepted title for a Muhammadan prince at least during the period *c.* 1000-1300 A.D. But this word may have even a longer history. It is not unlikely that it came to acquire this meaning from the middle of the 9th century A.D., when the Arab chiefs of Sind became practically independent and assumed the title of *Amīr*.¹

v2.c28.76

² *EI*, Vol. I, p. 135 and p. 129, line 28. On the identification of Vināyakapāla see *DHNI*, Vol. I, pp. 573 ff., 590 ff.

v2.c28.77

¹ *Ibid*, p. 147, lines 32-33.

v2.c28.78

² *TKA*, Bulak, 1874, Vol. IX, pp. 115-116.

v2.c28.79

³ The inscriptions of Kālañjara (mod. Kalinjar) mostly belong to a later period, the time of Madanavarman (c. 1128-63 A.D.) and Paramardi (c. 1167-1203 A.D.).

v2.c28.80

⁴ *IA*, Vol. XVI, p. 203, line 7.

v2.c28.81

⁵ Niẓāmud-Dīn says of this fort that 'it has no equal in the whole country of Hindustan for strength and impregnability;' *TA*, Trans., p. 14. Bīrūnī also refers to Kālañjar as one of the two 'most famous fortresses of India;' see his *Indica*, trans. by Sachau (Trübner), Vol. I, p. 202.

v2.c28.82

⁶ *IA*, Vol. XVI, p. 203.

v2.c28.83

¹ *EI*, Vol. I, p. 139, and p. 146, V. 55.

v2.c28.84

² *EI*, Vol. I, p. 145, V. 46.

v2.c28.85

³ *Kratha was a country in the peninsular portion of India, possibly near modern Berar. In the Mahābhārata (Bomb. Ed., II, Chap. 14, V. 21). Kratha-Kausikān are mentioned in connection with the Bhojas and Pāṇḍyas. Some identify Kratha-Kausika with Vidarbha. See GDI, p. 104.*

v2.c28.86

⁴ *EI*, Vol. I, p. 145, V. 45.

v2.c28.87

⁵ *Ibid*, p. 197, V. 3; also foot-note 62 on p. 203.

v2.c28.88

⁶ *KY*. Trans. from the Persian version by Reynolds, p. 456. *DHNI*, Vol. I, p. 608.

v2.c28.89

¹ *EI*, Vol. I, pp. 218 and 221, V. 17.

v2.c28.90

² *Encyclopaedia of Islam*, Vol. I, pp. 330 and 334.

v2.c28.91

³ S. Lane-Poole, *Mohammadan Coins*, Oxford, 1888, p. 18.

v2.c28.92

⁴ *Rājataraṅgiṇī*, VII, 53, 64, Stein's note on VII, 47-69 on pp. 270-71 in *Vol. I*, of his *Trans.* See also *DHNI*, Vol. I, pp. 95 ff., 135.

v2.c28.93

⁵ Wright, *Catalogue of the Coins of the Indian Museum*, Calcutta, Vol. II, Oxford, 1907, p. 17. Thomas, *The Chronicles of the Pathan Kings in Delhi*, p. 20, No. 18.

v2.c28.94

⁶ Wright, *op. cit.*, pp. 17-38; Thomas, *op. cit.*, pp. 15-127; also fn. 1, on pp. 50-51, *JASB*, Vol. XLIII, Part I, p. 108, plate X.

v2.c28.95

⁷ Thomas, *op. cit.*, pp. 31 and 100-03.

v2.c28.96

¹ See *DHNI*, Vol. I, chapter on Dynasties of Sind, pp. 91, fn. 3; *Elliot*, Vol. I, p. 36.

v2.c28.97

To return to our discussion of the epithet, "equalled even the powerful *Hamvīra*." The known dates of Dhaṅga (V.S. 1011-1059, *c.* 954-1002 A.D.) show that he was a contemporary of the Yamīnī Sulṭāns Sabuk-tigīn (A.D. 976-97), Ismā‘īl (A.D. 997) and Maḥmūd (A.D. 998-1030). It has been usually assumed by scholars that the *Hamvīra* referred to above must be identified with Sabuk-tigīn. They further assume that the Kalinjar ruler who according to Firishta 'supplied troops and money' to the Śāhi ruler Jayapāla must have been the Candella ruler Dhaṅga.² I have elsewhere tried to show that there are reasons to suspect the veracity of Firishta on this point.³ The silence of ‘Utbī, Ibn ul-Athīr, and Niẓām ud-Dīn appears to me to be rather significant. I have already pointed out that the last known date of Dhaṅga is V. S. 1059, and the earliest known date of his successors is 410 A.H.,⁴ so his death must have occurred sometime between *c.* 1002 and 1019 A.D.⁵ As he lived for more than a 'hundred autumns' (*Śaradāṁ śataṁ*)¹ it is not unlikely that he may have lived up to 1008 A.D., which date practically saw the downfall of the Sāhis. The defeat of Ānandapāla threw open the Ganges-Jumna valley to the inroads of the Yamīnī Amīrs. The shadow of the *Haṁvīra* fell beyond the waters of the Sutlej. The achievements of Maḥmūd only served to heighten the prestige of his arms. Under the circumstances, to compare Dhaṅga's prowess to that of the terrible *Haṁvīra* may have appeared to a late *praśastikāra*, writing at least a hundred years after that prince's death,² to be deserving of the highest reward. If Dhaṅga really fought and was defeated by the *Haṁvīra*, we should expect a discreet silence from the poets living at the court of his successors.

v2.c28.98

² *EI*, Vol. I, pp. 218-19; *IA*, 1908, p. 140, *JRAS*, 1909, pp. 276; *CHI*, Vol. III, p. 507.

v2.c28.99

³ See *DHNI*, Vol. I, pp. 83, 91-22, 597, etc.

v2.c28.100

⁴ The date of his grandson Vidyādhara; see *infra*, pp. 688 ff.

v2.c28.101

⁵ Kielhorn while editing the Khajraho inscription of Dhaṅga dated in V. S. 1059, renewed by Jayavarmadeva in V. S. 1173 remarked: "The date shows that Dhaṅga died between the Vikrama years 1055, the date of the Bengal Asiatic Society's copper-plate grant mentioned above (*IA*, Vol. XVI, pp. 202-04) and 1059." (*EI*, Vol. I, p. 139.) This has been accepted by other scholars, see *IA*, 1908, p. 141. But the inscription in question records in lines 32-3, V. 63: *Saṁvat 1059 Śrī-Kharjuravāhake Dhaṅga-deva-rājye deva-Śrī-Marakateśvarasya praśasti siddhā*. This clearly shows that Dhaṅga was alive in V.S. 1059. The fact that in line 29, V. 55 refers to the death of Dhaṅga only proves that the record in question was re-arranged and re-written when it was renewed after more than a hundred years by Jayavarmadeva in V. S. 1173. It is of course strange that the record 'should contain no allusions whatever to the successor of Dhaṅgadeva.' But this unusual character of the inscription remains unexplained even if we accept the theory of Kielhorn. To me it therefore seems that the epigraph does not give us any facts to fix the date of the death of Dhaṅga 'within very narrow limits of possible error.' It is clear that he did not die 'a little before 1002 A.D.' We only know that he died some time after A.D. 1002.

v2.c28.102

¹ *EI*, Vol. I, p. 146, V. 55.

v2.c28.103

² The Mahoba epigraph which contains this passage is unfortunately fragmentary. But in V. 28 it seems to refer to the son of Kīrtivarman (V. S. 1154). *EI*, Vol. I, p. 221.

v2.c28.104

The above discussion shows that Dhaṅga's reign was a long and distinguished one, probably unmarred by defeats at the hands of the Turuṣkas. The temples of Khajraho, which are regarded as 'the finest group of Hindu temples in Northern India,' and are usually referred to the 10th and 11th centuries, bear evidence to the splendour of the victories of the Candellas in the domain of peace.³ Some of these edifices certainly belong to the reign of Dhaṅga. The temple of Jinanātha, which contains an inscription of Dhaṅga's reign of the year 1011 recording a number of gifts by a Jain devotee, was probably built during that ruler's reign. The temple of Vaidyanātha, which contains an inscription of the year 1058 recording the building of a temple in honour of Vaidyanātha by one Kokkala of the *Gahapati* family, also appears to have been built during this reign.⁴ Another Khajraho inscription records the building of 'a magnificent temple for the god Sambhu, *Marakateśvara*, with two *liṅgas* one of emerald and the other of stone. This inscription is dated in the year 1059, and is now built into a wall of the temple of Visvanātha.¹

v2.c28.105

³ Cunningham, *ASR*, Vol. II, pp. 419 ff.; Fergusson, *History of Indian and Eastern Architecture*, 1910, Vol. II, pp. 49 ff., 140 ff.; *IA*, 1908, p. 63.

v2.c28.106

⁴ *EI*, Vol. I, pp. 147 ff.

v2.c28.107

¹ *Ibid*, pp. 137 ff.

v2.c28.108

The Mau stone inscription of the time of Madanavarman supplies us with the name of one of his ministers, named Prabhāsa, who was born in the lineage of Aṅgiras and Gautama Akṣapāda, the reputed founder of the *Nyāya* philosophy. He was, we are told, 'highly expert in the abstruse conduct of politics (*naya-prayoge gahane sudakṣaḥ*). After being '(duly) tried' (*sarv-opadhāśuddhi*) was appointed 'chief of all ministers' by Dhaṅga and king Gaṇḍa.²

v2.c28.109

² *EI*, Vol. I, p. 199, V. 21.

v2.c28.110

Dhaṅga after living for more than a hundred years at last "abandoned the body in the waters of the Ganges and the Yamunā, and entered into beatitude, closing the eyes, fixing his thoughts on Rudra, and muttering holy prayers." The following records and dates are known for his reign :³

v2.c28.111

³ > *Rakṣitvā kṣitim amburāśirasanām etām ananyāyatīṁ,* > *Jīvitvā śaradāṁ śataṁ samadhikaṁ Śrī-Dhaṅga-pṛthvīpatiḥ.* > *Rudraṁ mudrītalocanaḥ sa hṛdaye dhyāyañjapañjāhnavī-* > *Kālindyoh salile kalevara-parityāgād agānnirvṛtiṁ. (EI,* Vol. I, p. 146, V. 55.) Such acts of suicide are not unknown in Indian History; see *BG*, Vol. I, Part II, p. 442; *DHNI*, Vol. I, p. 347.

v2.c28.112

(1) *Khajraho stone-inscription.*—The stone which bears this inscription is said to have been discovered in the ruins at the base of the Lakṣmaṇa-temple at Khajraho. It consists of 28 lines and contains 49 verses in fluent and correct Sanskrit. It opens with adoration to the holy Vāsudeva' and then invokes the god Vaikuṇṭha. Next follows the account of the rise of the Candrātreyas from the Sage Atri. The genealogy is traced from Nannuka to Dhaṅga The inscription is mainly concerned with the *praśasti* of Yaśovarman, who constructed a splendid home of Viṣṇu and set up in it an image of Vaikuṇṭha, which he received from the *Hayapati* Devapāla (Vs. 23-43). The last 6 verses give an account of Dhaṅga; and then comes the date (V.) S. 1011 (A. D. 953-54) and the name of Vināyakapāla ‘protecting the earth.’ It ends with ‘adoration to the holy Vāsudeva! adoration to the sun.’ The inscription was composed by the *Kavi* Mādhava and written by ‘the *Karaṇika Gauḍa* Jaddha who was *Saṁskṛta-bhāṣāvidvas*.¹

v2.c28.113

¹ The inscription was first noticed by Cunningham in *ASR* (Vol. II, p. 426; Vol. XXI, p. 65), and a very small photolithograph of it was published, *ibid.* Vol. XXI, plate XVII. It was edited by Kielhorn in *EI*, Vol. I, pp. 123-135. The stone bearing this record is now built into the wall inside the entrance-porch of the temple of Lakṣmaṇa.

v2.c28.114

(2) *Khajraho Jain Temple inscription.*—It is carved on the left door-jamb of the temple of Jinanātha at Khajraho. It consists of 11 lines of Sanskrit, partly in prose and partly in verse. The inscription begins with ‘*oṁ*’ and then gives the date (V.) S. 1011 (A. D. 1055). It next records ‘a number of gifts made (probably in favour of the temple where the inscription is) by one Pāhilla,’ who was ‘held in honour by Dhā(a?)ṁga-rāja’ and was a devotee of the ‘lord of the Jinas.’ The gifts mainly consist of gardens (*vāṭikā*). Line 10 mentions the name of *Mahārāja-guru* Vāsavacandra.²

v2.c28.115

² The inscription was noticed by Cunningham in *ASR*, Vol. II, p. 433, and XXI, p. 67. He also published a photolithograph of it, Vol. XXI, plate XVI, No. J. It was first edited by R. L. Mitra in the *JASB*, Vol. XXXII, p. 279; re-edited by Kielhorn in *EI*, Vol. I, pp. 135-36.

v2.c28.116

(3) *Nanyaura (now Bengal Asiatic Society) grant.*—It contains 15 lines incised on one side of a single copper plate, and was found with another plate of Devavarman (dated in V. S. 1107) in a field at the village of Nanyaura, in the Panwari-Jaitpur Tahsīl of the Hamirpur District, U. P. There are no holes in the plate for rings, nor any seal attached to the plate The grant opens with the sign *Oṁ svasti* and then traces the genealogy of the donor from king Harṣadeva. We are told that *Pb.-M.-P.* Harṣadeva-*pādānudhyāta-Pb.-M.-P.* Yaśovarmanadeva-*pādānudhyāta-Pb.-M.-P.-Kālañjarādhipati*-Dhaṅga-deva in the (V.) year 1055 (A. D. 998), when there was an eclipse of the moon, granted from Kāśikā (Benares) the village of Yu(Cu?)llī attached to (*pratibaddha*) Uṣaravāha to the *Bhaṭṭa* Yaśodhara, an emigrant from Tarkārikā, whose ancestors had been settled at Dūrvāharā. The inscription ends with ‘*Śrī Dhaṅga*.’¹

v2.c28.117

¹ The plate was first edited by V. A. Smith in the *JASB*, Vol. XLVII, Part I, p. 84, and was re-edited by Kielhorn in *IA*, Vol. XVI, pp. 201-04. It is now in the Bengal Asiatic Society.

v2.c28.118

(4) *Khajraho stone inscription of Kokkala.*—The slab which bears this inscription is said to have been found at the temple of Vaidyanātha at Khajraho. The record contains 22 lines. It is a *praśasti* of a Grahapati family which originally came from Padmāvatī (mod. Narwar, Gwalior State). The inscription opens with ‘*Oṁ namaḥ Śivāya*,’ and then invokes Śiva in several verses under the names Vaidyanātha, Śaṅkara, and Śarva. Then the genealogy of Kokkala is traced from Yaśobala or Atiyaśobala, of Padmāvatī. In lines 19-21 is recorded the erection of a temple of Vaidyanātha and a number of buildings for pious Brāhmans by Kokkala at Khajraho. The date (V.) *Saṁvat* 1008 (A. D. 1000-01) is given in the last line.²

v2.c28.119

² Noticed by Cunningham in his *ASR*; see Vol. XXI, p. 66, and plate XIX. Edited by Kielhorn in *EI*, Vol. 1, pp. 147-52. It is now built into the wall of the entrance of the porch of the temple of Visvanātha.

v2.c28.120

(5) *Khajraho stone inscription.*—It was discovered in a temple at Khajraho. It consists of 33 lines. The record opens with *Oṁ namaḥ Śivāya*, and then praises Śiva (Rudra, Digambara, Śūladhara, Maheśvara), Bhāratī, and Gaṇeśa. Next follows the usual story of the origin of the Candrātreyas. Then comes the regular genealogy of the family from Nannuka to Dhaṅga. In lines 48-51 we are told that the latter erected a magnificent temple for the god Sambhu. “He also distributed great quantities of gold and established in connection with the temple dwellings for pious Brāhmans to whom donations were made of land, grain, money, and cows.” (Vs. 52-54.) The record was composed when the illustrious priest of the royal household Yaśodhara was acting in the office of *Dharmādhikāra*.³

v2.c28.121

³ This Yaśodhara is most probably to be identified with the donee of the Nanyaura plate, *op. cit.*, No. 3.

v2.c28.122

Lines 32-33 give the date as follows: "The (V.) year 1059 (A. D. 1001-02); at the glorious Kharjuravāhaka, in the *rājya* of the illustrious king Dhaṅga, this *praśasti* of the god *Marakateśvara* was completed." The record was composed by the poet Rāma, grandson of the *kavi* Nandana of the *Sāvara* (*Śābara*)-*vaṁśa* and a resident of Tarkārikā. It was written by the *Kāyastha* Yaśaḥpāla, and engraved by Siṁha. The temple was built by the *Sūtradhāra* Chicca.¹

v2.c28.123

¹ The inscription was first most carelessly edited by Sutherland in *JASB*, Vol. III, pp. 159-84. It was re-edited by Kielhorn in the *NKGWG*, 1886, pp. 441-62; finally edited by the same scholar, *EI*, Vol. 1, pp. 137-47. The inscription was probably re-arranged and re-written in V. S. 1173, when it was renewed by Jayavarman. This explains the occurrence of the description of Dhaṅga's death in V. 55, while the date of his record is given in V. 63. The inscription is now built into the wall on the right side of the temple of Viśvanātha.

v2.c28.124

Dhaṅga was succeeded by his son Gaṇḍa sometime between 1001-02 A. D., Dhaṅga's last known date, and 1019 A. D., the first recorded date of his grandson Vidyādhara. So far we have not discovered any inscription of this prince. His name is only mentioned in some late records of the Candellas which were composed at least a hundred years after his death. The following statements regarding his reign can be collected from these epigraphs.

First 133 of 422 passages. Read on

Linked texts (41)