Pramāṇa · Search · About · API

Indian Epigraphy c14: Chapter VI. Indian Epigraphy Abroad

Indian Epigraphy c14

Language
English

Contents

Chapter VI. Indian Epigraphy Abroad

c14.2

With the spread of Indian culture over various countries of Asia, the Indian languages and scripts were introduced into many regions especially of Central and South-East Asia. Epigraphic and literary records in Sanskrit, Prakrit and the regional languages, written in Kharoṣṭhī, Brāhmī and the derivatives of the latter have been found in Central Asia which was regarded as forming a part of the ancient Uttarāpatha division of Bhāratavarṣa. The current alphabets of most of the South-East Asian countries are derivatives of Brāhmī. The earliest inscriptions of these regions, with the exception of Burma and Ceylon, are in Sanskrit. While Prakrit and Pali inscriptions are more numerous in Ceylon, Sanskrit and Pali were both used in early Burmese epigraphs.

c14.3

There is a general tendency among scholars to describe the alphabet used in the early inscriptions discovered in the countries of Indonesia and Indochina as Pallava-Grantha and to speak of the expansion of Indo-Aryan culture in those territories from the Coromandel Coast during the rule of the Pallavas.¹ These views are however wrong. In the first place, the alphabet, used in the Pallava inscriptions from the 4th to the 6th century A.D. as well as in most of the early epigraphs of the lands beyond the Bay of Bengal, cannot be called Pallava-Grantha. It is really the Late Brāhmī common to the southern and western parts of India. Secondly, the earliest inscriptions of Indochina and Indonesia are written in verse like the records of the Early Kadambas of the Kannaḍa-speaking area and unlike those of the Early Pallavas.²

c14.4

¹ See, e.g., JRASB, Vol. I, 1935, pp. 1 ff.

c14.5

² It is interesting to note that the name Kaṅgavarman found in the Phnom Bayang inscription of Bhavavarman II of Kambuja reminds us of such personal and geographical names as Kaṅgavarman, Koṅgaṇi, Koṅgudeśa, Koṅgi-nagara found in the southern part of the Kannaḍa-speaking area and its neighbourhood. See Journ. G. Ind. S.n., Vol. V, p. 156; Suc. Sāt., pp. 249, 252; etc.

c14.6

Thirdly, the box-headed alphabet used in certain epigraphs of king Bhadravarman of Campā (South Annam)¹ is not a characteristic of Pallava inscriptions but is generally found in the epigraphs of Central India and the Upper Deccan as well as in the Kannaḍa-speaking western regions of South India. Fourthly, the Śaka era introduced in Indochina and Indonesia about the 6th century A.D. was never used in the inscriptions of the Pallavas, the Cālukyas of Bādāmi (from the middle of the 6th to that of the 8th century A.D.) being responsible for the association of the name Śaka with the said era and the popularity of its use especially in the present Kannaḍa and Marāṭhī-speaking areas. Fifthly, a number of South-East Asian inscriptions, particularly those of the early ruling families of Arakan (Burma) and of the Śailendras of Śrīvijaya (Palembang in Sumatra), are written in the Late Brāhmī and Siddhamātṛkā alphabets of Eastern India.² Sixthly, geographical names like Irāvatī (the river Irawady in Burma; cf. the old name of the Rāvi in the Punjab), Ayuthia (city in Siam; cf. the ancient city of Ayodhyā in the Fyzabad District, U.P.), Candrabhāgā and Gomatī (canals or streams mentioned in the Tugu inscription of Pūrṇavarman of Java; cf. the celebrated North Indian rivers of these names), Campā-pura (the old capital of South Annam; cf. the same name of the ancient capital of Aṅga or East Bihar), Kambuja (ancient name of Cambodia; cf. the name of the Kamboja people of North-Western Bhāratavarṣa), etc.,³ appear to associate the people responsible for coining the names with various areas of Northern India even though some of them may be merely Indianised forms of local names. Seventhly, there are some records in these countries which definitely point to a contact especially with East India among territories outside South India.

c14.7

¹ Cf. *JRASB*, op. cit., Plate I, fig. 1 (facing p. 10).

c14.8

² *Ep. Ind.*, Vol. XXXII, pp. 103 ff.; cf. Chatterji and Chakravarti, *India and Java*, Part II, p. 5.

c14.9

³ Some other similar names of towns and countries common to India (excluding the Far South) and the lands beyond the Bay of Bengal are Kauśāmbi, Dvārāvatī, Mathurā, Kaliṅga, Gandhāra, etc. Cf. R. C. Majumdar, *Champa*, p. xxiii. Note also the representation of *b* by *v*.

c14.10

The above facts suggest that the responsibility for the spread of Indian culture to the lands beyond the Bay of Bengal cannot be assigned exclusively to the Pallava dominions on the Coromandel Coast. As a matter of fact, the peoples of various parts of India, especially of the coastal regions—both eastern and western, contributed to the spread, the contributions of the Coromandel Coast being only a part of the whole and those of the western areas of the Deccan as also of East India being quite considerable.

c14.11

The name-ending *varman* generally found in the royal names of these regions is not only found in the Pallava family but is also noticed in the ruling dynasties flourishing in other parts of India, e.g. the Aulikaras of Mandasor, the Śālaṅkāyanas of Veṅgī and the Kadambas of Banavāsi, all of whom began to rule from the 4th century A.D., as well as the Gaṅgas of Kaliṅganagara (from the end of the 5th century) and the Maukharis of Gayā (5th century). Most of the Kaliṅga dynasties of the 5th century, e.g. the Māṭharas and Pitṛbhaktas, had royal names ending in *varman*. It has to be noticed that none of the Varman kings of the lands beyond the Bay of Bengal appear to have flourished earlier than the close of the 5th century A. D.

c14.12

In this connection, attention may be drawn to the traditions regarding the Indian colonisers in the said countries, e.g., the Ceylonese story of the conquest of the island by Vijaya, a great-grandson of the king of Vaṅga (then Southern Bengal); the foundation of Ligor ascribed to one of Aśoka’s descendants, who fled from Magadha (South Bihar) and embarked on a vessel at Dantapura lying near modern Srikakulam in the ancient Kaliṅga country comprising wide areas of the coast lands of Orissa and Andhra Pradesh; the story preserved in the Javanese chronicles as well as in the traditions of many other islands regarding the colonisation of the lands in question by the people of Kaliṅga (generally called Kling);¹ a tradition of Pegu regarding the colonisation of the Irawady delta and the adjoining coast lands of Burma by the people of the lower valleys of the Kṛṣṇā and the Godāvarī in the eastern coast lands of the Deccan; the story in the Javanese chronicles regarding the colonisation of the island by a prince of Gujarāt in 75 A.D.; etc.

c14.13

¹ The inscriptions of Andhra Pradesh and Orissa are, however, written in prose as in the case of those of the Coromandel Coast while the Śaka era was introduced in Orissa at a late date. The Orissan alphabet was not of the southern type.

c14.14

The same story is told by what we know about Indian trade with Indochina and Indonesia, vaguely called Suvarṇabhūmi or ‘the Land of Gold’ in Indian literature. Thus the Jātaka stories show how the people of Bengal, Bihar and Eastern U.P. came down to the port of Tāmralipti in the present Midnapur District in South-West Bengal, whence they sailed for the eastern islands. The Periplus and Ptolemey’s Geography refer to the city of Gaṅgā near the principal mouth of the river of that name as the main port on the Bengal coast. There were similar ports in the Puri-Ganjam region of the Kaliṅga country as well as in the region about the mouths of the Kṛṣṇā and the Godāvarī. The Greek sources speak of certain ports in the coast lands of the Tamil and Kannaḍa-speaking areas. Bhṛgukaccha (modern Broach or Bharuch) and Sūrpāraka (modern Sopara in the Thana District) were the most famous ports in the coastal area of Western India.¹ There is some indication that all these ports had trade relations with the lands beyond the Bay of Bengal.

c14.15

¹ Cf. R. C. Majumdar, Champa, pp. xi-xii.

1. Central Asia

c14.s1.1

Aśoka’s inscriptions discovered in the Peshawar and Hazara Districts of West Pakistan are written in the Prakrit language and Kharoṣṭhī alphabet, the latter being an Indian modification of the Aramaic script introduced in the region during the Achaemenian occupation. But an important inscription of the Maurya king discovered in Afghanistan is the Kandahar bilingual epigraph which is in two versions, one in Greek and the other in Aramaic.² Among later records from the Afghanistan area written in Prakrit and Kharoṣṭhī, mention may be made of the inscriptions of the times of Menander and Vijayamitra found in the Bajaur tribal territory and the vase inscription of Huviṣka’s time from Khawat (Wardak) about 30 miles to the west of Kabul.¹

c14.s1.2

² Journal Asiatique, 1958, pp. 1 ff.; Ep. Ind., Vol. XXXIV, pp. 1 ff. A few other Aramaic inscriptions have also been ascribed to Aśoka.

c14.s1.3

¹ Select Inscriptions, pp. 102 ff., 153-54.

c14.s1.4

But, before the 6th century A.D., the use of Kharoṣṭhī died not only out of the north-western regions of the Indian subcontinent but also out of Afghanistan. The latest Kharoṣṭhī inscriptions in India are two small Taxila records of about the 5th century A.D., though we have a Brāhmī inscription of about the 3rd century A. D. from the same area.² The Gardez (Afghanistan) Vināyaka image inscription in Sanskrit, now in the Dargah Pir Ratan Nath at Kabul, may be mentioned in this connection. It is written in the Late Brāhmī alphabet of the 6th century A.D. and is stated to have been installed by the Śāhī king Khiṅgāla in the 8th year of his reign.³ A late medieval inscription in Nāgarī characters is noticed on the left side of the Central Mihrab of the rock-cut chamber near Chahil Zina (literally ‘forty steps’) in the vicinity of old Kandahar.⁴

c14.s1.5

² Cf. Marshall, Taxila, Vol. I, pp. 274 ff.; Ep. Ind., Vol. XXX, pp. 59 ff.

c14.s1.6

³ Cf. Ramachandran and Sharma, Archaeological Reconnaissance in Afghanistan (Preliminary Report of the Indian Archaeological Delegation 1956), 1956, pp. 1-2. See East and West, December 1958, pp. 327-28; Ep. Ind., Vol. XXXV, pp. 44 ff.

c14.s1.7

⁴ Ramachandran and Sharma, op. cit., p. 19. For similar records from Baku (Azerbaijan), see A. R. Ep., 1956-57, Nos. B 547 ff. We have seen the photograph of an inscription of the same type from Samarkand (Ujbekistan).

c14.s1.8

A large number of manuscripts and documents written in Kharoṣṭhī with ink on wedge-shaped and rectangular tablets of wood have been found in Chinese Turkestan. They are usually assigned to dates ranging from the 3rd century A.D. There are also manuscripts like that of the Prakrit Dhammapada which is written in Kharoṣṭhī on birch-bark sheets and is assigned to the 2nd century A.D. Kharoṣṭhī continued to be used in the said area till the 7th century A.D. side by side with Brāhmī which seems to have been introduced in Central Asia during the Gupta age. Among Sanskrit manuscripts in Brāhmī on palm-leaf, paper and birch-bark, reference may be made to Kumāralāta’s *Kalpanāmaṇḍitikā* and the *Udānavarga*, both of which are assignable on palaeographical grounds to the 4th or 5th century A.D., and the Bower manuscripts, the date of which is about a century later. Other Sanskrit manuscripts in Brāhmī, discovered in Central Asia, include fragments of Aśvaghoṣa’s drama entitled *Śāriputraprakaraṇa*. From about the 7th century A.D., we have records in the local dialects of Central Asia written in the derivatives of Brāhmī.¹

c14.s1.9

¹ See my *Indian Palaeography*, Chapter II, Section 6, and Chapter III, Section 5.

c14.s1.10

The earliest epigraphs in Nepal are the Rummindei and Nigali Sagar pillar inscriptions of the Maurya emperor Aśoka (3rd century B.C.) written in the Early Brāhmī alphabet and Prakrit language.² Early Licchavi epigraphs like the Changunarayan pillar inscription (464 A.D.) of Mānadeva are written in Sanskrit and the Late Brāhmī of East India, while the characters of the Sanskrit records of the later members of the Licchavi family are Siddhamātṛkā.³ Though most of the early medieval records of Nepal belonging to later dates, both epigraphic and literary, are written in Gauḍīya or the East Indian derivative of Siddhamātṛkā, some Buddhist manuscripts are written in Bhaikṣukī, while Buddhist Dhāraṇīs, etc., in Sanskrit were often written in a derivative of the Siddhamātṛkā alphabet called Rañjā or Rañjanā which was also used for that purpose in Tibet.⁴ The Tibetan alphabet itself is a slightly modified form of Siddhamātṛkā. Buddhist Dhāraṇīs were written in China and Japan in the Siddhamātṛkā alphabet till quite recent times.

c14.s1.11

² *Select Inscriptions*, pp. 70-71.

c14.s1.12

³ See Sylvain Lévi, *Le Nepal*, Part III, pp. 10 ff. and Plates. Cf. R. Gnoli, *Nepalese Inscriptions in Gupta Characters*, 1956.

c14.s1.13

⁴ See my *Indian Palaeography*, Chapter V, Section 3.

c14.s1.14

Early medieval manuscripts of a large number of works copied in East India have been found in Nepal and Tibet. Of these, the Tibetan collection contains mostly Buddhist works, while both Buddhist and Brahmanical works are found in the Nepalese collection in large numbers. The discovery of many manuscripts in Nepal, which were copied in the dominions and during the reigns of the Pāla kings of Bengal and Bihar, is explained by the fact that, during the early days of the Muslim occupation of Eastern India, many scholars, afraid of Turkish persecution, fled to the north, especially to the Nepal region, together with the books in their family libraries. There is also evidence to show that professional copyists of one area were often employed for copying manuscripts in the neighbouring regions.¹

c14.s1.15

¹ Cf. IHQ, Vol. XXX, pp. 382-87.

2. Ceylon

c14.s2.1

Among the countries of South-East Asia, Ceylon’s cultural contact with India is the oldest. According to an old Ceylonese tradition, Prince Vijaya, son of the king of Siṃhapura in Rāḍha (in South-West Bengal; or less probably Lāṭa in Gujarat) and a great-grandson of the king of Vaṅga (then South Bengal), colonised Ceylon about the beginning of the 5th century B.C. About the middle of the 3rd century B.C. the Maurya emperor Aśoka of Pāṭaliputra (near modern Patna) is known to have maintained friendly relations with Tāmraparṇī or Ceylon. He is also stated to have sent to that country a Buddhist mission which converted Devānāmpriya Tiṣya, king of the island, to Buddhism together with his subjects. Ceylon’s contact with the Tamil-speaking region of South India, just beyond the Palk Strait, has been very close throughout the ages and, during the major portion of the 11th century A.D., wide areas of the island formed parts of the empire of the Coḷas of Tañjavūr. It is therefore interesting to note that the Ceylonese language belongs to the Aryan (and not the Dravidian) group. Old Ceylonese traditions appear to suggest that the peoples of Eastern and Western India were mainly responsible for the spread of Aryanism in that land.² The Ceylonese royal house often contracted matrimonial alliance with Indian ruling families such as the Pāṇḍyas of the South and the Kaliṅgas of the East.

c14.s2.2

² See CHI, Vol. I, pp. 605-606.

c14.s2.3

The close contact of Ceylon with non-Tamil India in pre-Christian times seems to be indicated by another fact. We know that the early independent rulers of India were satisfied with the simpler title Rājan while the Indo-Greek kings of North-Western Bhāratavarṣa introduced the royal title Rājā mahān or Mahārāja, which was essentially a translation of the old Persian title Kṣāyaθiya vazṛka enjoyed by the Achaemenian monarchs, about the middle of the 2nd century B. C. The earliest indigenous Indian monarch assuming the title Mahārāja is king Khāravela of Kaliṅga who flourished about the close of the 1st century B.C., while the same title appears in Ceylonese epigraphs which may be somewhat earlier in date.¹ This seems to suggest that the title was borrowed by Ceylon from Western or North-Western Bhāratavarṣa which came under the influence of the Greeks. Some of these early Ceylonese records have to be read from right to left in the Kharoṣṭhī fashion.² This also seems to point to Ceylon's contact with the Kharoṣṭhī-using areas of the said part of India. The letter ḷ in early Ceylonese epigraphs is not idicated by a diacritically marked l as used in the Early Brāhmī alphabet of the Tamil areas but by ḍ endowed with a diacritical mark as known from other parts of India.

c14.s2.4

¹ Cf. Select Inscriptions, pp. 206 ff., 232-33.

c14.s2.5

² Ibid., p. 233.

c14.s2.6

The story of the introduction of Indianism in Ceylon is different from that of its spread in Indochina and Indonesia which came under the influence of Indian merchants, adventurers and priests about the beginning of the Christian era. In many of these countries Brahmanical Hinduism not only spread earlier than Buddhism but Śaivism seems to have been the dominating religious faith throughout the early period. This was not the case with Ceylon where Buddhism was always the predominant faith as it is now.

c14.s2.7

Sanskrit inscriptions are rare in Ceylon, the more popular Indian languages there being Prakrit and Pali. The Ruvanavalisāya pillar inscription in Late Brāhmī of about the 5th century A.D. and the Kuchchaveli rock inscription in a developed form of the same alphabet assignable to a date about the 7th century A.D. are two of the few early Ceylonese records in Sanskrit.¹ The Gārandigala inscription of about the first half of the 8th century A.D. is one of the early epigraphs in Ceylonese while the Sanskrit inscription from the Jetavanārāma is written in the Siddhamātṛkā alphabet of Northern India and may be assigned on palaeographical grounds to a date in the 9th century A.D.² Many Tamil inscriptions, especially of the time of the Coḷa occupation of Ceylon during the 11th century A.D., have been discovered in the island.³

c14.s2.8

¹ *Ep. Zeyl.*, Vol. III, pp. 120 ff., 158 ff.

c14.s2.9

² Ibid., pp. 195 ff.; Vol. I, pp. 1 ff. The earliest record dated in the Buddha-nirvāṇa era is the Anuradhapura inscription of year 941 (=28th regnal year of Buddhadāsa’s son Upatisya), Kuja-vāra, Pūrṇimānta-Pauṣa-badi 1=16th December 396 A. D. (*Univ. Cey. Rev.*, Vol. XVIII, pp. 129 ff.).

c14.s2.10

³ Cf. *SII*, Vol. IV, Nos. 1388 ff.

3. Burma, Siam (Thailand) and Malay Peninsula

c14.s3.1

Among the early inscriptions of Burma, there are epigraphs in Sanskrit and in Pali, records in the latter being mostly Buddhist tracts. The simple Sanskrit inscriptions in verse, belonging to the Candras of Arakan and assignable to the 6th century A.D., resemble in style the records of king Mūlavarman of Borneo. It is interesting to note that, unlike Mūlavarman’s inscriptions, the alphabet of these records is the Late Brāhmī of Eastern India, while the said style seems to have been borrowed from South Indian epigraphs like those of the Early Kadambas of Banavāsi in the Kannaḍa-speaking region. Ānandacandra’s inscription (8th century A. D.) from the same region of Burma is a rare instance of a Sanskrit epigraph of the praśasti type found in that country, elaborate eulogies in Sanskrit verse written in ornate kāvya style being common in the other lands of Indochina and Indonesia. The inscription of Ānandacandra is written in the Siddhamātṛkā alphabet of Eastern India.⁴

c14.s3.2

⁴ See *Ep. Ind.*, Vol. XXXII, pp. 103 ff.; *Sel. Ins.*, pp. 462 ff.

c14.s3.3

Most of the early inscriptions found elsewhere in Burma are however written in the Late Brāhmī alphabet of West and South India. There is also evidence to prove the contact of Tamil-speaking merchants with Burma during the early medieval period. But South-East Burma seems to have closer ties with the people of East India. The Śaka era was not popular in Burmese epigraphy as it is in the inscriptions of the other regions of South-East Asia.

c14.s3.4

The earliest inscriptions in the language of the Mons (Talaings), who lived in the Irawady basin with Hanthawady (Haṃsāvatī, i.e. Pegu) and Thaton (Sudharmapura) as their chief cities, were found on a stone slab discovered at Lopburi in Siam.¹ They are written in characters of about the 6th or 7th century A.D. and point to the cultural contact between the two countries. A large number of Pyu inscriptions assignable to dates between the 6th and 8th centuries A. D. were found at Hmawsa (Prome) and Halin in the Shwebo District.² The Myazedi (Myinkaba, Pagan) inscription of the 12th century A.D. contains 4 different versions, viz. Pali, Burmese, Mon and Pyu.³ A Pagan inscription of about the 13th century A.D. is written partly in the Sanskrit language and Grantha characters and partly in the Tamil language and alphabet and records the grant of certain gifts by a Vaiṣṇava devotee of Cranganore in Malabar in favour of a Viṣṇu temple at Pagan.⁴

c14.s3.5

¹ Rep. Dir. Arch. Surv., Burma, 1956, pp. 19 ff.

c14.s3.6

² Cf. Ep. Ind., Vol. XII, pp. 127 ff.; ASI, A.R., 1911-12, p. 142; 1926-27, p. 172; 1927-28, p. 145; 1929-30, p. 182; ASB, A.R., 1942, p. 22.

c14.s3.7

³ Ep. Birm., Vol. I, pp. 1 ff.

c14.s3.8

⁴ Ep. Ind., Vol. V, pp. 101 ff.; Vol. VII, pp. 197-98.

c14.s3.9

An early Sanskrit epigraph discovered in Siam is the Ban Bungke inscription (886 A. D.) of king Indravarman of Cambodia.⁵ The Sukhotai (Sukhodaya) inscriptions (Śaka 1214) of king Rām kham-heng of Siam are written in the Siamese language and early Siamese characters.⁶

c14.s3.10

⁵ BEFEO, Vol. XXII, p. 63.

c14.s3.11

⁶ D. Diringer, The Alphabet, p. 416.

c14.s3.12

The earliest inscriptions discovered in the Malay Peninsula are Buddhistic in nature.¹ A prose endorsement in one of them, besides the Buddhist creed, speaks of *Mahānāvika* (i.e. ‘the great sailor’) Buddhagupta of Raktamṛttikā which was a locality near modern Murshidabad in South-West Bengal.² The alphabet of these records is however the Late Brāhmī of West and South India assignable to the 5th or 6th century A.D.

c14.s3.13

¹ Chatterji and Chakravarti, *India and Java*, Part II, pp. 6-7.

c14.s3.14

² *Select Inscriptions*, p. 365; cf. *Ep. Ind.*, Vol. XXVII, p. 2.

c14.s3.15

The earliest epigraph dated in the Śaka era in the Malay Peninsula is the Vieng Sa (Ligor, South of the Bay of Bandon) inscription of Śaka 697 (775 A. D.).³

c14.s3.16

³ Chatterji and Chakravarti, op. cit., pp. 40 ff.

4. Indonesia

c14.s4.1

The earliest inscriptions of Indonesia are written in Sanskrit verse. Most of the early medieval records of the area are in the ornate *kāvya* style; but their size is not big as in the case of Cambodian inscriptions. Java developed an indigenous literature originated from an intimate acquaintance with Indian literary tradition. Although the people of the area are now Muhammadans, the Indian element is still prominent in their culture. The people of Bali may be regarded as Hindu in both religion and culture.

c14.s4.2

The inscriptions of Mūlavarman from Kutei in Borneo are written in *Anuṣṭubh* stanzas in the style of some of the inscriptions of the Early Kadambas of Banavasi, their alphabet being Late Brāhmī similar to that found in Buddhagupta’s inscription discovered in the Malay Peninsula and referred to above.⁴ Pūrṇavarman’s inscriptions from Java are written in a similar alphabet and style, there being one stanza in *Sragdharā* besides others in *Anuṣṭubh*. They may be assigned to the 6th century A.D.⁵

c14.s4.3

⁴ *Select Inscriptions*, pp. 446-47; Chatterji and Chakravarti, op. cit., pp. 8 ff.

c14.s4.4

⁵ *Select Inscriptions*, pp. 468 ff.; Chatterji and Chakravarti, op. cit., p. 20 ff.

c14.s4.5

Epigraphic records of Indonesia continue in an unbroken series from the 5th or 6th century down to the end of the Indo-Javanese period about the beginning of the 16th century. But the number of Sanskrit inscriptions found in Malay Peninsula and Indonesia is much smaller than that of those discovered in Cambodia and Annam. The Indonesian inscriptions generally record pious donations or the building of temples. They usually inform us as to when and by whom a temple or monastery or image was constructed or a donation made. When a gift of land was the subject of an inscription, its boundaries were indicated, and, when privileges were bestowed their nature was written down. Usually the inscriptions give some information about the kings in whose regnal reckoning they were dated as well as about the royal officers associated with the documents together with facts throwing light on the religious, administrative and social condition of the country.

c14.s4.6

The inscriptions are generally written on stone, either slabs or rocks. But there are some epigraphs on plates of copper, silver or gold. The earlier inscriptions are in Sanskrit in the Late Brāhmī alphabet of West and South India or its local modification; but later epigraphs are mostly in the Kavi or old Javanese language and the Kavi script which was a derivative of Brāhmī developed in the land itself. The Kavi language is an admixture of Sanskrit and the local Polynesian dialect. The Śailendra kings of Śrīvijaya used the Siddhamātṛkā alphabet of East India and the Sanskrit language for their inscriptions.

c14.s4.7

The use of the Śaka era became popular in Indonesian epigraphy, its earliest use being noticed in inscriptions of the 8th century. It was probably introduced by the merchants of the Western part of the Deccan. The earliest record of Java dated in the Śaka era is the Kelurak inscription¹ of Śaka 700 (778 A.D.). The Minto stone inscription² of Śaka 876 (954 A.D.) is in the Kavi language but has two Sanskrit stanzas at the beginning. The Pereng stone inscription (Śaka 785-863 A.D.), written in the Kavi language, contains five Sanskrit stanzas in the Āryā metre.¹ Similar is the case with the Singasari inscription (915 A.D.) of king Daṣka written in both Sanskrit and Kavi.²

c14.s4.8

¹ India and Java, II, pp. 44 ff.

c14.s4.9

² Ibid., p. 48.

c14.s4.10

¹ Cf. ibid., p. 48, note 6.

c14.s4.11

² Kern, *V. G.*, Vol. IV, pp. 279 ff.

5. Cambodia

c14.s5.1

Unlike Java, Cambodia does not possess a literature influenced by Indian literary tradition. But the Sanskrit inscriptions, dating from the 5th to the 14th century, found all over Cambodia, exhibit a very flourishing state of Sanskrit learning. The number of such inscriptions written in ornate kāvya style is the largest in Cambodia than in any other land of South-East Asia and many of them are quite voluminous. The use of the Śaka era and of the decimal system in numbers is first noticed in Cambodia respectively in the 6th and the 7th century. The first of them appears to have been introduced in this region by the merchants of the western areas of the Deccan, and the second probably by those of the southern areas of the Gujarātī-speaking region. In India, the decimal system is noticed for the first time in the date of the Sankheda (in the old Baroda State in Gujarat) copper-plate inscription of 594 A.D.³

c14.s5.2

³ Bhandarkar’s List, No. 1205.

c14.s5.3

The study of Indo-Cambodian epigraphy began in 1879 with the decipherment of some Sanskrit records by H. Kern from the estampages prepared by J. Harmand. A collection of estampages of such epigraphs made by E. Aymonier was published under the name *Inscriptions Sanscrites du Cambodge* by A. Barth and A. Bergaigne, in two Parts with a volume of Plates, the first Part edited by Barth appearing in 1885 and the second edited by Bergaigne with notes by Barth in 1893. The inscriptions in the native Khmer language in Aymonier’s collection were translated and summarised by himself in his *Le Cambodge*, Vols. I-III (1901-04). A large number of inscriptions in Sanskrit and Khmer, discovered at later dates were published by L. Finot, G. Coedès and others in the *Bulletin de l'Ecole Francaise d'Extreme-Orient* appearing originally from Hanoi. Coedès has also published a large number of Sanskrit and Khmer inscriptions in several volumes (1937 onwards) of his *Collection de textes et documents sur l'Indochine* with several volumes of Plates.

c14.s5.4

Late Brāhmī of West and South India, which was introduced in Cambodia, gradually underwent modification in the same way as in different parts of India. The orthography of the Sanskrit epigraphs of Cambodia exhibit the substitution of *ḍ* for *d* and *ṇ* for *n* and vice versa and sometimes even of *ṇḍ* for *nd*. The use of *ṅ* for *anusvāra*, the change of visarga before a sibilant to the same sibilant, the occasional reduplication of a consonant before and after *r* as well as before *v*, the use of *tv* for *ttv* and of *jihvāmūlīya* and *upadhmānīya* for visarga before the guttural and labial surds, the occasional use of *candrabindu* for *anusvāra* and the rarity of cases in which *v* is distinguished from *b*, etc., are some of the orthographical features of the Sanskrit inscriptions of Cambodia.¹

c14.s5.5

¹ R. C. Majumdar, *Inscriptions of Kambuja*, pp. xiv ff.

c14.s5.6

The language of the records is mostly correct Sanskrit, irregularities and mistakes which are few being probably due to the scribe or engraver rather than the composer. The records, which are generally incised on stone are mostly engraved with great care. The majority of the inscriptions are written in beautiful language in the *kāvya* style exhibiting the authors' thorough acquaintance with the metres and the rules and conventions of Sanskrit rhetorics and prosody. The Vat Thipedi inscription² (Śaka 832=910 A. D.) of king Iśvaravarman II is composed in the flawless Gauḍī style of composition, and it has been suggested that the author was an inhabitant of Gauḍa in East India or lived there for a long time. A large number of the inscriptions contain 50 stanzas or more, while some contain more than 100 stanzas. The Mebon (Śaka 874) and Pre Rup (Śaka 883) inscriptions³ of Rājendravarman contain no less than 218 and 298 stanzas respectively.

c14.s5.7

² Ibid., pp. 161 ff.

c14.s5.8

³ Ibid., pp. 193 ff., 232 ff.

c14.s5.9

The epigraphs are usually *praśastis* extolling the king and recording details of endowments created in favour of religious institutions. Unlike Indian inscriptions of the *praśasti* type, the Cambodian eulogies of the kings often give, besides other details of historical importance, the date of accession of the reigning monarch as well as his ancestors. But, as in the case of Indian epigraphs, the eulogies usually contain a great deal of what is conventional.

c14.s5.10

The details of religious endowments, which are the subject of most of these records, supply the names of deities (fairly large in number) worshipped in the country together with long lists of articles of daily or seasonal worship; the utensils and other necessaries of the temples including their personnel; the accounts of temple properties both movable and immovable and the method of their management and various other matters connected with temples and throwing light on the religious, social and economic life of the people. There are remains of a large number of temples all over Cambodia, the inscriptions relating only to some of them.

c14.s5.11

The Cambodian inscriptions are mostly of Śaivite character, although there are some Vaiṣṇavite and a few Buddhistic records. The inscriptions show how Indian culture was adopted in Cambodia by some of the local people who were semi-savages roaming about naked about the beginning of the Christian era when the Indians first came into contact with them. Some interesting facts about life in Cambodia, as revealed by the inscriptions, are the worship of Śiva, Viṣṇu and the Buddha side by side; the existence of different religious sects like the Bhāgavatas, Pāñcarātras and Pāśupatas; the adoption of the caste system dominated by Brāhmaṇas versed in the Vedas, Vedāṅgas and Upavedas as well as of Sanskritic names by the kings, noblemen and common people; the popularity of the Tantric religion and of well-known Tantric texts; the existence of Devadāsīs in the temples and of priests of the same family sometimes conducting religious rites of the royal families for centuries in succession; the creation of new castes by the king and admission of new members to a caste such as that of the goldsmiths; the popularity of the study of the works of Indian authors like Pāṇini, Patañjali, Manu, Vātsyāyana, Viśālākṣa, Pravarasena, Mayūra, Guṇāḍhya and Suśruta. King Yaśovarman is stated to have composed a commentary on Patañjali’s *Mahābhāṣya*, while one of his ministers was an expert in the *Horā-śāstra*.¹ The same king is known to have founded 100 *āśramas* which were a sort of convents or monasteries. The Śaivas, Vaiṣṇavas and Buddhists had their own *āśramas* which were religious centres radiating Indian culture throughout Cambodia.

c14.s5.12

¹ Ibid. p. xvii.

c14.s5.13

Some of the inscriptions of Yaśovarman, who flourished about the end of the 9th and the beginning of the 10th century A.D., are written in the Siddhamātṛkā alphabet while some of his other records giving regulations about *āśramas* are both in the said alphabet and in the Cambodian derivative of Brāhmī. This fact suggests that there were fresh arrivals apparently from North or East India, who were not familiar with the latter alphabet. Contact with South India during the early medieval period seems to be indicated by the Prasat Kandol Dom inscription² of king Indravarman who flourished about the end of the 9th century, according to which the king’s *guru* Śivasoma studied the scriptures at the feet of Bhagavat Śaṅkara, apparently the great Indian philosopher of the same name who was born in the western coast of South India.

c14.s5.14

² Ibid., pp. 57 ff.

c14.s5.15

The Śaka era appears in Cambodia in inscriptions like one of Śaka 531 from Prasat Ak Yom while the decimal system of writing numbers is first noticed in a pillar inscription of Śaka 605 (683 A. D.) from Sambaur.³ The Sambaur pillar inscription contains 21 lines written in the Khmer language which was used in Cambodian epigraphy from about this time side by side with Sanskrit.

c14.s5.16

³ See ibid., pp. 7, 564. For the Roban Romas inscription of Saka 520, see *BEJEO*, Vol. XLIII, pp. 1 ff.

6. South Annam

c14.s6.1

The number of Sanskrit inscriptions discovered in South Annam (ancient Campā) is fairly large; but the size of most of them is not very big. Many of the epigraphs are written in verse; but there are some written either in prose or in a mixture of prose and verse. They are generally on stone stelae, although some of them are engraved on rocks, stone images and metalic plates. The palaeography of the earlier records is similar to that of the Cambodian epigraphs. The Dong Duong inscription of Indravarman II, dated Śaka 797 (875 A.D.), is a fairly big record engraved on the four faces of a stelae containing respectively 24, 24, 23 and 31 lines.¹

c14.s6.2

¹ R. C. Majumdar, *Campā*, Part III, pp. 74 ff.

c14.s6.3

A collection of the Sanskrit inscriptions of Campā was prepared by A. Bergaigne in 1888. It was published after his death by Barth in 1893 in the *Notices et Extraits de Manuscrits de la Bibliothèque Nationale et autres Bibliothèques*, Vol. XXVII, Part 1, Fasc. 2. A large number of inscriptions in the Cham language were edited by Aymonier in the *Journal Asiatique*, 1891. Inscriptions in Sanskrit and Cham, discovered at later dates, have mostly been published in the *Bulletin de l'Ecole Francaise d' Extreme-Orient* by scholars like L. Finot, M. Huber, G. Coedès and others.

c14.s6.4

The earliest inscription discovered in Campā is the Vocañh epigraph of a king of the Śrīmāra family who flourished in the 4th or 5th century A. D.² The Mi-son and Cho Dinh inscriptions of king Bhadravarman, who flourished in the 5th or 6th century A. D., are written in the box-headed alphabet.³ The Śaka era began to appear in the records of this area about the second half of the 6th century A.D.⁴

c14.s6.5

² Cf. *IHQ*., Vol. XVI, pp. 484 ff.; Vol. XVII, pp. 107 ff; *Journ. As*., 1953, pp. 477 ff. Some scholars assign the inscription to an earlier date.

c14.s6.6

³ See *BEFEO*, Vol. XXXII, 1932, pp. 127 ff. and Table; *JRASB*, Vol. I, 1935, Plate facing p. 10.

c14.s6.7

⁴ Cf. R. C. Majumdar, *Campā*, Part III, pp. 9 ff.

Linked texts (6)