A similar case which shows the emergence of another Hinduized state in Indonesia is provided by the oldest dated inscription from Central Java, found in Canggal, a village in the region of Kedu (Sarkar 1971-72:I, 15-24). This Sanskrit inscription tells us about the foundation of a lingga, the phallic symbol of God Śiva, by a King Sañjaya in AD 732. It also mentions his father, named Sanna or Sannāha, who had “in the fullness of time, gone to enjoy happiness in the heaven which is the accumulated results (of his meritorious deeds)”.[2] Since Sannāha, despite its spelling, is not a Sanskrit word, it seems reasonable to suppose that it was a Sanskritized Javanese name. We have a similar case in the name of Sañjaya’s successor. His Javanese name, as it appears in the Old Javanese Mantyasih I inscription of AD 907, was Panangkaran, but in the Sanskrit Kalasan inscription dated 778 his name appears in its Sanskritized form as Paṇaṃkaraṇa (Sarkar 1971-72:II, 64-81; I, 34-40).
Some geographical names may also appear in Sanskrit forms. Thus the name Tārumā, mentioned in the fifth century inscriptions found in western Java, may have been a Sanskritized form of an indigenous name related to tarum, “indigo” (Gonda 1973:77). Another example is Yava which according to Gonda “forms a vexed question”. The name Yavadvīpa, “Island of Barley”, for Java and Sumatra, probably even for Kalimantan, was already known in early Sanskrit literary works (Wheatley 1961:177-179). But since barley is grown in neither island, Gonda (1973:348-350) argues that the original name was Jawa, which was a word for “Indonesian, indigenous” and could have applied to both Java and Sumatra, to their inhabitants, products, languages, etc. Indian navigators and colonizers interpreted the word, by way of popular etymology, as yava-, “abundance of barley” and named the island accordingly. Then, under the Sanskritizing tradition, the new name “Yava” returned to Java as we find it in the Canggal inscription.[3]
Since Sannāha apparently did not have a Sanskrit name it seems very likely that Sañjaya, like Aśwawarman of the Kutai inscription, was the first of his line to adopt the śivaite cult, establish a “new” kingdom, and assume a Sanskrit name. Like Aśwawarman he was regarded as the founder of the dynasty by later generations, as is evident from the Mantyasih I charter. This is an Old Javanese inscription issued by King Balitung in AD 907, in which Sañjaya’s name appears at the head of a list of eight successive rulers of Mataram. Significantly, in this charter he is simply entitled rakai Mataram sang ratu Sañjaya, while all the other rulers are recorded with a more exalted title, Śrīmahārāja, e.g. Śrīmahārāja Panangkaran, Śrīmahārāja Panggumulan, etc.
However, Sanna himself was also a ruler. This is clear from the Canggal inscription which tells us that he, “by means of conciliation and gift, ruled the subjects in a proper way, out of attachment, just like a father (taking care of) the child from his very birth” (verse 8c). In the context of the organization of communities in ancient Java, the choice of the simile “like a father (taking care of) the child” to describe the relationship between the ruler and the subjects seems to be very apt. As is known from the Old Javanese inscriptions, the smallest indigenous community in ancient Java was the wanua (“village, settlement”). Its inhabitants, especially those born there, were called anak wanua or, from the end of the tenth century, anak thāni (“children of the village”). They were probably the descendants of the original founders of the village concerned. The wanua was governed by a board of elders presided over by the first among equals, the rāma, the primary meaning of which was “father” (van Naerssen 1977:37; de Casparis 1990:62, note 6). By using such a simile, the author of the inscription might have intended to indicate that Sanna was a rāma, or, if he were not one, that he governed his subjects in the manner proper to a rāma.
The latter seems to have been more likely. We know from the inscription Mantyasih I that his son Sañjaya was a rakai, one level higher than a rāma in the social hierarchy of ancient Java. A rakai was the head of a territorial unit called watak or watĕk (“group”), which consisted of several wanua. The rakai-ship must have had its origin in prehistoric times. As van Naerssen has argued, with the introduction of wet-rice cultivation a more complicated system of irrigation was required and co-operation by the several wanua which depended on water of the same river or its tributaries therefore became necessary. Such an irrigation system needed a head whose authority reached beyond that of a single wanua, and so a leader, rakai, was elected from among the rāma (van Naerssen 1977:37). It appears from the inscriptions, however, that in most cases the villages belonging to the same watĕk had become so dispersed that they no longer formed a contiguous territorial unit. By the tenth century a watĕk might have comprised villages that were located in different parts of central and/or eastern Java, and the names of the well-known watĕk such as Hino, Halu and Sirikan gradually lost their original territorial connotation.[4]
It has been suggested that the word rakai or rake was derived from raka, meaning “older brother” (Stutterheim 1933:165; van Naerssen 1977:37). Such an explanation, however, is now no longer acceptable. It has been pointed out that the word raka “never occurs in the inscriptions with the meaning of ‘older brother’, nor with any other meaning” (Barret Jones 1984:93), and the form raka i is not in fact found in any original inscriptions (de Casparis 1990:56). Moreover, considering that ranking based on age is a very important principle among Austronesian peoples, one would expect that a kinship term indicating an older age group than “father” would be used to denote a position higher than rāma. It seems more likely, therefore, that the word rakai and variants such as rake, reke and raki, is related to ra-(ka)ki or ra-kya, the primary meaning of which is “elder, grandfather”.[5]
The highest authority in the autochthonous hierarchical system of ancient Java was the ratu. As far as we can see from the inscriptions, while rakai apparently no longer had direct relationships with the wanua from which they or their ancestors came, a ratu still maintained close ties with his own watĕk, as is evident from the use of his rakai title beside his ratu title. Thus, as we have seen, Sañjaya is recorded as rakai Mataram sang ratu Sañjaya in the Mantyasih I charter. Since Sannāha, as we shall see below, was also a ratu, it seems reasonable to suppose that he himself was also a rakai like his son.
The word ratu, or its cognates, is found in many languages which belong to the Austronesian language family with a variety of meanings, such as Tagalog dāto (“high priest”), ’Toba Batak datu (“sorcerer”), Malay datok (“head of a kin group”), and Fijian ratu (“title of rank before names of males who are chiefs”). Based on these various meanings, Blust (1980:216-217) suggests that the Proto-Austronesian * datu had at least four components of meaning, namely: (1) political leader, chief; (2) priest, custodian and administrator of customary law, medical practitioner; (3) aristocrat, noble; and (4) ancestor, grandfather, elder.
Thus the Old Javanese ratu, like rāma and rake (or rakai), may have developed from a kinship term meaning “ancestor, grandfather” to mean “political leader, chief”. In terms of ranking rāma (“father”) was head of the smallest unit, the wanua; rake (“elder, grandfather”) was head of the watĕk; and ratu (“grandfather, ancestor”) was the highest in the hierarchal system.
Since Sañjaya was given Sanskrit titles meaning “king”, i.e. narapati and rāja, in the Canggal inscription (in verses 1 and 11 respectively), it is obvious that the author of the inscription, and most likely local Sanskrit scholars in general, considered ratu as the equivalent of Sanskrit rāja, “king”. In this meaning (sang) ratu also occurs frequently in the Old Balinese inscriptions, especially during the reign of Ugrasena (915-39) (Damais 1949:29; Goris 1954:II, 296). Dātu also occurs in the Old Malay inscriptions of Śrīwijaya, probably with the meaning of “governor of provinces”, but kadātuan, significantly, was used for “royal residence” (de Casparis 1956:38, 345), resembling Old Javanese kaḍatwan (“royal residence, kingdom”) and Modern Javanese kraton.
Like Sañjaya, Sannāha was also entitled rāja in the Canggal inscription (verse 8). So, he too must have been a ratu during his lifetime. In fact, he may have been the descendant of a long line of ratu, as the inscription specifically says that he was rājogrodagrajanmā , that is “a king of a very noble lineage”. Likewise, Kuṇḍungga of the Kutai inscription must also have been a ratu, or whatever word was used in Kutai at that time to denote “political leader, chief”, for in the inscription he was called a narendra, which is synonymous with Sañjaya’s designation as narapati (“lord of men”).
We may thus conclude that both Aśwawarman in Kutai and Sañjaya in Mataram did not actually found new kingdoms or principalities, but merely transformed the old established kaḍatwan into new rājya. It is moreover evident from the great number of official titles occurring in the Old Javanese inscriptions, the overwhelming majority of which are non-Indian and non-Sanskritic (Sarkar 1971-72:I, xix), that the new kingdom was largely a continuation of an older one.
The establishment of these new kingdoms obviously took place only after their rulers had decided to adopt Hinduism. Although it is often said that one is Hindu only by birth, the process of “Hinduization”, which is sometimes given the more general term “Indianization” or “Sanskritization” (Hall 1981:12; Coedès, 1968:15-16; Mabbett 1977) seems to have occurred throughout the long history of India and still continues to occur (Srinivas 1966:1; Coedès 1968:25). It has been pointed out by Srinivas (1989:63) in this context that it was the second varṇa, the kṣatriya, that “seems to have been the one most open, accommodating all kinds of groups, indigenous as well as alien, the only necessary qualification being the effective possession of political power”. For the elevation of a native chief to the level of kṣatriya the Brahmanic rite of vrātyastoma must have been performed by Brahmans. Accordingly, in areas where there was no established Brahman the chief either had to import some from outside — offering them gifts of land and other inducements — or even create them himself from amongst ambitious local groups (Coedès 1968:24; Srinivas 1989:63).
Perhaps in order to show that one had the necessary qualifications — the effective possession of political power — it seems to have been mandatory to announce one’s conquest of surrounding areas in inscriptions. Thus we read, for instance, that Mūlawarman had “conquered other kings in the battlefield, [and] made them his tributaries” (Kutai inscription), and that Sañjaya had overthrown many “circles of feudal lords” (Canggal inscription).
As far as the need for the Brahmans to perform the necessary rites is concerned, the Kutai inscriptions explicitly tell us that the sacrifices held by Mūlawarman were performed by Brahmans who “had come there”. At this early stage of the Sanskritization of the Indonesian archipelago they were likely to have been Indian Brahmans, but it is also possible that they were Austronesians who had acquired a priestly education in India and who came from other parts of the country for special occasions. Whatever the case, they must have come there on the invitation of the king, who then rewarded them with religious gifts, land and great wealth (up to twenty thousand and probably forty thousand cows are mentioned in the inscriptions — though the numbers were most likely exaggerated). Some may have left soon after the events, but a number of them must have decided to stay, forming the nucleus of a growing number of indigenous learned men. A Chinese record tells us that in the kingdom of P’an-p’an (fifth century AD) “are numerous brahmans come from India in search of wealth. They are in high favour with the King” (Wheatley 1961:49).
It is clear from the above examples of the Hinduization of Kutai and Mataram that Hinduism — or Buddhism in cases where the rulers, for one reason or another, preferred the Buddhist monks — was disseminated by Brahmans invited there by the local rulers, and that these religions spread principally among the limited circles of members of the royal courts. It was, as Coedès (1968:33) puts it, “essentially an aristocratic religion which was not designed for the masses”. And, like the “new” kingdom which still retained many of the essential parts of the old system, the new cult also contained many autochthonous religious beliefs. Thus one could point out, for instance, that behind the new Indian deva there was the ancient Javanese hyang, Old Javanese for “deity” (indeed the new deities were given the honorific hyang, e.g. hyang Śiwa, hyang Wiṣśṇu); that śiwa was easily accepted as the highest God because, as the son-in-law of Mount Himalaya, he could be related to ancient worship of the Great Mountain; that the Javanese caṇḍi (“temples”) can be identified as successors of the terraced religious sites of prehistoric times; or that the best known Javanese Buddhist monument, Borobudur, can only be understood properly in the light of ancestor worship (de Casparis 1950:188; Holt 1967:35-38).
It is clear, then, that the adoption of the Indian concept of kingship and various forms of Sanskritic culture and Indian religions did not lead to fundamental changes throughout the Indonesian political and social order[6] The sheen of these imported religions and cultural forms, to borrow a much quoted metaphor, “is a thin and flaking glaze”, under which the whole of the old indigenous forms has continued to exist (van Leur 1967:95).
Nevertheless, significant changes did occur as the result of the penetration of the Sanskritic culture into the western parts of the Austronesian world. There seems to be little doubt that Sanskritization provided the necessary ingredients to enable the local cultures, especially those that had had more intensive contacts for a lengthy period of time, to produce cultural manifestations distinctly different from those of other Austronesian people who had no direct contacts with Sanskritic culture. Without the infusion of the Indian conception of royalty, it is extremely doubtful whether local polities with the relatively small-scale systems of political integration and ranking which were the common features of prehistoric Austronesian societies, could have developed into “true states with specialised bureaucracies and the powers to maintain allegiance by force” (Bellwood 1985:146-148). And without the foundation of such states it is very unlikely that religious monuments on the scale of the universally acclaimed temples of Borobudur and Prambanan — or even the smaller ones, the remains of which are scattered all over central and eastern Java as well as in many other places outside Java — could ever have been built.