Indian Epigraphy c6.s5: 5. Demerits of Epigraphic Evidence
Indian Epigraphy c6.s5
- Language
- English
5. Demerits of Epigraphic Evidence
c6.s5.1
Although inscriptions have contributed largely to the reconstruction of the lost history of ancient and early medieval India, their evidence is not free from defects. Some of these are briefly discussed below.
c6.s5.2
It has to be admitted that only some of the inscriptions can be regarded as truly historical documents. No doubt many of them deal with historical events and personages; but history is often shadowed in them by poetical, eulogistic and conventional elements. Most records therefore give the impression that references to historical events in them are incidental. Their evidence is thus often indirect and leaves many things to be surmised and inferred. It is only rarely that we get in inscriptions fairly full and satisfactory information regarding an event or a person. As already pointed out, the authors of inscriptions always suffered from a limitation of space and their treatment of history was never elaborate. Sometimes the genealogical section of the inscriptions quotes the names of past rulers with the barest detail or without any details at all.
c6.s5.3
We have seen above how the use of popular eras in dating documents was unknown in ancient India, how even royal records were dated in the regnal reckoning of the rulers of the land, and how eras were introduced and popularised in the country by foreign rulers. But, even after the popularity of the eras had been well established in wide areas of the country, documents of many parts were not endowed with verifiable dates. Moreover, a large number of records of all ages and of all parts of the land are altogether undated. In many cases, the details of the dates quoted in dated records contain errors owing to mistakes in the calculation of astronomical data either by the persons responsible for the records or by the astronomers on whom they depended.¹ Although, therefore, dated inscriptions have solved many of the knotty problems of chronology in ancient and medieval Indian history, a number of problems still remain unsolved owing to the said defects of epigraphic records. In the absence of a date in a known era, an epigraph can only be approximately referred to a period, but not to a definite date.
c6.s5.4
¹ Cf. *IHQ*, Vol. XXVIII, pp. 342 ff.
c6.s5.5
A tendency is often noticed in ruling families from the early medieval age to fabricate a respectable genealogy. The process began with the elaboration of the genealogy on the basis of the *gotra* name of a royal family. The Pallavas of Kāñcī claimed to have belonged to the Bhāradvāja-gotra which offered them the opportunity to trace their descent, through a number of mythical and imaginary personages, from the sage Bharadvāja, famous in the epics and the Purāṇas, and through him to the god Brahman, the creator.² Among royal families of the early medieval period, it was a fashion to claim descent from the solar, lunar, Yādava and other dynasties, well-known from ancient Indian literature.³ Many of the rulers claiming such respectable descent belonged really to aboriginal families. Legends were often fabricated to explain the name of a royal family, sometimes with reference to an eponymous hero.⁴
c6.s5.6
² Cf. *SII*, Vol. I, p. 9; *Ep. Ind.*, Vol. XVIII, p. 150; etc.
c6.s5.7
³ Note the cases of the Gurjara-Pratīhāras, Rāṣṭrakūṭas and Imperial Gaṅgas claiming descent respectively from the solar, Yādava and lunar dynasties.
c6.s5.8
⁴ Note the cases of families like Pallava, Paramāra, Pratīhāra, etc.
c6.s5.9
In some cases, the fabricated element in the genealogy can be easily detected. In the records of Vajrahasta III (1038-70 A.D.) and Rājarāja I (1070-78 A.D.) and also in the earlier epigraphs of Anantavarman Coḍagaṅga (1078-1147 A.D.), Vajrahasta III of the Imperial Gaṅga dynasty of Orissa is represented as the grandson’s great-grandson of a king named Guṇamahārṇava (Guṇārṇava) of the Ātreya-gotra. But, in the later records of the family, including inscriptions belonging to the later years of Anantavarman Coḍagaṅga himself, the origin of the family is traced to the god Viṣṇu through his navel-born son Brahman, his mind-born son Atri, his eye-born son Candra (the Moon), his son Budha, his descendants Purūravas, Āyus, Nahuṣa, Yayāti and Turvasu—all known from epic and Puranic literature, and a host of apparently imaginary personages standing between Turvasu and Guṇārṇava who was the progenitor of the family according to the earlier account, but is here represented as Guṇārṇava II. Names of Kolāhala, founder of Kolāhalapura in the Gaṅgavāḍi-viṣaya in Mysore, and his successors, some of whom are said to have migrated to Kaliṅga, are inserted before the reference to Guṇārṇava obviously in an attempt to connect the Gaṅgas of Orissa with those of Mysore, even though the latter claimed the Kāṇvāyana-gotra. The genealogy was fabricated on the basis of the *gotra* claimed by the Eastern Gaṅgas; but it also exhibits their eagerness to claim descent from the lunar dynasty and some connection with the god Viṣṇu, to whom Anantavarman Coḍagaṅga transferred his allegiance from Śiva during the latter part of his career.¹
c6.s5.10
¹ See *Ep. Ind.*, Vol. XXVIII, pp. 237 ff.
c6.s5.11
The authors of the *praśastis* were mostly attached to royal courts and eager to exaggerate the achievements of their patrons and the latters’ ancestors. Such exaggeration often mars considerably the value of inscriptions of the *praśasti* type as a source of history. A few illustrations will make the point sufficiently clear.
c6.s5.12
According to the Mahakuta pillar inscription,² the Cālukya king Kirtivarman I (566-98 A.D.) of Bādāmi defeated the rulers of Vaṅga, Aṅga, Kaliṅga, Vaṭṭūra, Magadha, Madraka, Kerala, Gaṅga, Mūṣaka, Pāṇḍya, Dramila, Choḷiya, Āluka and Vaijayantī. It is impossible to believe that Kīrtivarman I actually came into conflict with such far off lands as Vaṅga, Aṅga, Magadha and Madraka, especially in view of the fact that, in the Aihole praśasti of his own son Pulakeśin II who can hardly be expected to have concealed any of his father’s glorious achievements, Kīrtivarman I is credited with military success only against the Nalas, Mauryas and Kadambas, all of whom were neighbours of the Cālukyas of Bādāmi. The real basis of the tall claim made on behalf of Kīrtivarman I in the Mahakuta pillar inscription seems to be imagination.
c6.s5.13
² *Ind. Ant.*, Vol. XIX, p. 7.
c6.s5.14
The Candella king Dhaṅga (c. 950-1002 A.D.) is described in a stanza of an inscription at Khajuraho¹ (Chhatarpur District, Madhya Pradesh), which implies that the monarch crushed many kings, including the rulers of Kāñcī, Andhra, Rāḍhā and Aṅga, and had the queens of all the defeated rulers imprisoned in his capital. There is little doubt that the claim is an exaggeration. In the first place, it is doubtful if Dhaṅga at all came into collision with all the four kings mentioned in the record, even if the rulers of Aṅga and Rāḍhā are regarded as merely viceroys of the contemporary Pāla emperor of Eastern India. Secondly, even if Dhaṅga actually fought with the four kings, it is more doubtful that he succeeded in defeating all of them. Thirdly, even if he was victorious in all the four cases, it is difficult to believe that he succeeded in carrying off the queens of all his adversaries. Fourthly, even granting that Dhaṅga actually captured the queens of his enemies, he would normally be expected to have placed the ladies in his harem or in the charge of his favourite subordinates rather than in the prison.²
c6.s5.15
¹ Ep. Ind., Vol. I, p. 145.
c6.s5.16
² See my Studies in the Geography of Ancient and Medieval India, pp. 1 ff.
c6.s5.17
Gross exaggerations are, however, less conspicuous in the description of kings found in records of the earlier period. For this reason, in spite of the fact that there is always an amount of exaggeration in the royal praśastis composed by the court poets of Indian monarchs, the earlier the king is the greater is our reliance on his claims. Vague claims are generally less reliable than definite statements involving the mention of the personal names of adversaries. Whatever be the nature and amount of exaggeration, there is undoubtedly a very considerable amount of truth in the claims put forward in records like the Allahabad pillar inscription¹ of Samudragupta and the Tirumalai rock inscription² of Rājendra-coḷa. The manner in which the personal names of adversaries are mentioned in epigraphs like the Deopada inscription³ of Vijayasena (c. 1095-1158 A.D.) of Bengal appears, however, to be too poetic and artificial and therefore suspicious. But there is no doubt that the kings so named were contemporaries of the Sena monarch.
c6.s5.18
¹ CII, Vol. III, pp. 1 ff.
c6.s5.19
² Ep. Ind., Vol. IX, p. 299.
c6.s5.20
³ Ibid., Vol. I, p. 305.
c6.s5.21
In some cases, the poets are found more eager to display their skill in rhetoric rather than in the accuracy of their statements. The Candella king Yaśovarman (middle of the 10th century A.D.), who was a feudatory of the Gurjara Pratihāra emperor of Kanauj, is described in a Khajuraho inscription⁴ as a subduer of the Gauḍas, Khaśas, Kośalas, Kaśmīras, Mithilās, Mālavas, Cedis, Kurus and Gurjaras, though during the period in question neither were Kośala, Mithilā and Kuru separate states nor was the Candella king likely to have had anything to do with them. It seems that the lure of alliteration (cf. kośalaḥ Kośalānām, śithilita-Mithilaḥ, Kuru-taruṣu marut, etc.) carried the poet far away from historical accuracy.⁵
c6.s5.22
⁴ Ibid., p. 126.
c6.s5.23
⁵ Note the poet's eagerness to exhibit his knowledge of the Kāmaśāstra in a verse ascribed to Mayūra (Tripathi, History of Kanauj, p. 121), in which the words aṅga, kuntala, cola, madhyadeśa and kāñcī are used in the description of the king's enjoyment of the earth, conceived as a lady, in two senses; e.g. aṅga refers to the seizure of ‘the body’ [of the lady] as well as of the Aṅga country. Cf. Bhār. Vid., Vol. XV. No. 3, p. 26.
c6.s5.24
The authors of the praśastis were reluctant to take notice of the defeat and discomfiture of their patrons and their ancestors. Such unpalatable facts were often completely suppressed, and at times they were only vaguely and euphemistically referred to. Powerful feudatories were sometimes described deliberately in ambiguous language that can be interpreted as applicable both to a paramount sovereign and to a subordinate ruler.¹ In medieval inscriptions, the evidence of which can be checked with that of various sources, we have instances not only of an indecisive battle described as a victory but even of a defeat represented as a victory. Rāṇā Kumbhā of Mewar fought an indecisive battle with the Sulṭān of Malwa, but went back to Chitor to raise a kīrti-stambha bearing a praśasti² in commemoration of his victory over the Sulṭān! Rāṇā Udayasiṃha is represented in an inscription³ as having shadowed the glory of Akbar, even though wide areas of his dominions, including the capital city of Chitor, had passed to the Mughals.
c6.s5.25
¹ Cf. Ep. Ind., Vol. XXVIII, p. 186.
c6.s5.26
² Bhandarkar’s List, No. 797.
c6.s5.27
³ Ep. Ind., Vol. XXIV, p. 68.
c6.s5.28
The description of kings in the praśastis often contain claims that are conventional and therefore of little historical value. One of these conventions is the representation of an imperial ruler either as the conqueror or as the ruler of the entire ‘earth’ meaning the Cakravarti-kṣetra or ‘the sphere of influence of a paramount ruler.’⁴ This ‘earth’ was conceived as identical with ancient Bhāratavarṣa; but sometimes it was regarded as conterminous with either Āryāvarta (for North Indian kings) or Dakṣiṇāpatha (for South Indian monarchs). We know that the Pāla king Devapāla (9th century) ruled over Bengal and Bihar; but he is sometimes described in inscriptions as the ruler of the entire land bounded by the Himalayas in the north, Setubandha-Rāmeśvara in the south, the Bay of Bengal in the east and the Arabian Sea in the west, although sometimes the southern boundary is given as the Vindhyas instead of Setubandha. In the Nagpur praśasti⁵ of the Paramāras, king Lakṣmavarman (c. 1087-97 A. D.) is represented as the conqueror of the Gauḍas, Aṅgas and Kaliṅgas in the east, the Coḷas and Pāṇḍyas in the south in the region of the river Tāmraparṇī and the bridge of Rāma, and the Turuṣkas (Muhammadans) on the banks of the Vaṅkṣu (Oxus) in the north. There is little doubt that the description is a modification of Raghu’s digvijaya as described in Kālidāsa’s Raghuvamśa. It is impossible to think that the Paramāra king had any opportunity to encamp on the banks of the Oxus and to have come into contact with the Turuṣkas there. This exploit reminds us of the mythological Raghu’s advance as far as the banks of the same river inhabited by the Hūṇas in the age of Kālidāsa. Similar is the case often with titles assumed by some medieval Indian monarchs. King Kapilendra (c. 1435-66 A.D.) of Orissa called himself Gauḍeśvara-Navakoṭi-Karṇāṭa-Kalavargeśvara, i.e. lord of Gauḍa as well as of Karṇāṭa-nine-crores and Kalavarga (Gulbarga). The claim is, however, tall enough even for the mighty Kapilendra. But it is more interesting to note that petty Oriya rulers, who claimed to be his descendants, continued to assume the same title till recent times. But such instances are not too many, although the same description is often noticed to have been applied to royal residences situated in different localities.¹
c6.s5.29
⁴ See JRASB, Vol. V, 1939, pp. 407 ff; my Studies in the Geography of Ancient and Medieval India, pp. 2 ff.
c6.s5.30
⁵ Ep. Ind., Vol. II, pp. 186-88.
c6.s5.31
¹ Cf. the description of the jayaskandhāvāras in the charters of the Pālas.
c6.s5.32
We often see that a mere present from the ruler of a distant land could be pompously represented by the court poet as tribute, and contact of any kind with a king put up as his subordination to the poet’s patron or his ancestors.²
c6.s5.33
² Cf. The Classical Age, ed. Majumdar, p. 245 and note 3.
c6.s5.34
Often illiterate and semi-literate stone-cutters or goldsmiths were entrusted with the task of engraving records on stone or copper plates, and this fact accounts for the numerous errors noticed in a large number of epigraphs, especially those engraved on behalf of private individuals. There are many instances of indifferently drafted and badly engraved records even among royal charters, particularly those issued by minor ruling families. Mistakes committed by the scribes and engravers of inscriptions often lead to difficulties of interpretation and to controversies. Some of the controversies are due to ambiguity of expression from which many epigraphic passages suffer. As we have already noticed, in several cases such ambiguity was purposely resorted to in the description of feudatories aspiring for an independent status.
c6.s5.35
There are often discrepancies in the same account quoted in different records of a family. Thus the date of the accession of the Eastern Gaṅga king Kāmārṇava (1147-56 A.D.) is quoted correctly in the Dasgoba plates¹ of 1198 A.D. and the Nagari plates² of 1230-31 A.D. as the Śaka year counted by *nanda* (9), *ṛtu* (6), *vyoman* (0) and *candra* (1), which according to the *vāmagati* principle would yield Śaka 1069, corresponding to 1147 A.D. But all the later charters of the family, so far discovered, wrongly read *veda* (4) in the place of *nanda* (9) and thus make the date five years earlier. This mistake must have crept into the later records of the Gaṅgas exactly as errors usually creep into manuscripts in the process of copying.
c6.s5.36
¹ *Ep. Ind.*, Vol. XXXI, pp. 259 ff.
c6.s5.37
² *Ibid.*, Vol. XXVIII, p. 242.
Linked texts (2)
- Raghuvaṃśa — Kālidāsa, Sanskrit
- Mahākūṭa pillar inscription (No. 276) — Kannada