sulphur-scented levels, they were walking over an ancient place of power and pilgrimage.
It would be easy to overlook Java. A slender slip of land, six hundred miles long and less than a quarter of that across, it rests beneath the equator at the point where the southern arc of the Archipelago bears away eastwards towards distant New Guinea. It is a fraction of the size of Sumatra or Borneo, and is some distance from the crucial shipping junction in the Straits of Melaka. At a glance, then, it is not the most likely candidate for the cradle of history. But peer more closely at Javanese geography, and something becomes apparent. The Archipelago-long string of volcanos that is moderately spaced down the length of Sumatra and scattered, island by island, through Nusa Tenggara bunches up dramatically here, with a legion of fiery peaks marching in tight formation along the entire length of Java. Dieng is just one of dozens of active or dormant Javanese volcanoes. They provide a spine for the island, and though their capacity for sporadic violence gives the place an uneasy edge, the eruptions have doused the soils with nutrients. The peaks snatch at passing weather systems and squeeze out their moisture, and the bowls of land between them are well-watered. Java is one of the most fertile places on earth. What’s more, its mellow northern littoral is indented with safe river-mouth anchorages and brushed by easy trade winds.
There were already Indianised states here well before the rise of Srivijaya in Sumatra. They are glimpsed in the passing comments of Chinese scribes and travellers. In 412 CE, Faxian—the wandering monk whose story would inspire Yijing almost three centuries later—stopped by on his return from India in a West Java state that seems to have been called Holotan. As a good Buddhist, Faxian was none too approving of the state of religious affairs he encountered in Java: ‘Heretic Brahmans flourish there, and the Buddha-dharma hardly deserves mentioning’, he noted.
Later another state—this one called Tarumanagara—grew up in the same place, and there were others, hinted at in Chinese chronicles and traced in scattered stone pillars, marked with the wriggling worm-casts of the Pallava script. But the relics of these earliest Javanese states are remarkably thin on the ground, for the inhabitants built their homes and palaces of wood and thatch, and in the hot, wet climate these materials would last little more than a single monsoon once a kingdom had collapsed.
But by the seventh century, a new state had appeared on the northern coast of Central Java. When the Chinese heard of this polity they noted its name as ‘Ho-Ling’. This may have been a corruption of ‘Kalingga’, or perhaps ‘Areng’, but whatever it was called it marked the beginning of a relay of royal realms in Java that has continued all the way to the present day. And under the Ho-Ling aegis an epic tradition began of religious architecture in a medium more permanent than mere wood.
Java’s earliest stone temples sprang up on the Dieng Plateau in the second half of the seventh century. People from the surrounding hills had probably been making offerings to ancestral spirits at the steaming sulphur vents and cool caves here for many centuries, but these new places of worship were sophisticated miniatures of Indian influence and local innovation. The builders cleared spaces amongst the stunted trees, and raised carven blocks into squat but finely formed towers. Soon there were some two hundred temples scattered across the plateau and the surrounding hillsides. They were modest structures that were rarely more than twenty-five feet (eight metres) high. But their design set the pattern for a coming epoch: the leering kala ogres above the portals; the narrow access steps; the three levels of construction that symbolised the worlds of the mortals, the enlightened and the gods. And soon the royals of the rice lands below the Dieng eyrie would take these conventions and inflate them to a truly epic scale.
Down the steep slopes from Dieng along trails beetling back and forth through the forest, under the ribbed flanks of the Sindoro and Sumbing volcanoes and southeast across smoky foothills and deep green ravines, a returning pilgrim would come to the lush levels of Kedu. This bowl of low, well-watered land was hemmed with forested hills. To the east rose the hulk of Gunung Merbabu and its mighty twin, Merapi, angriest of all the volcanoes of Central Java. To the north an easy route led between the mountains towards the ports of the north coast, and snaking over the plain came the Progo River, a band of pale water which ran southwards, out of Kedu itself and across a triangle of rich, level land spreading south from Merapi. This was all fabulously fecund country. Palms stood in long ranks and the hillsides were thick with forest. The Progo was fed by myriad smaller streams, churning around bleached boulders and driving deep clefts into the limitless black soil, and the well-watered plots gave out a ceaseless cycle of crops across the seasons. It was a true land of milk and honey, and it would soon become the enduring cradle of Javanese civilisation.
By the early eighth century the main political action in Java seems to have shifted from the north coast to Kedu and the wider basin of the Progo. In 732, a local raja had his craftsmen raise an inscribed stone on a hilltop south of Merapi. It told of his rule over ‘a wonderful island beyond compare called Yava’. This king was called Sanjaya, a title he would pass on to a storied dynasty, and the general term for the area of Java over which he claimed control was Mataram, a name that would come to have such a hold on the Javanese imagination that it would eventually be revived by a line of Muslim sultans nearly a thousand years later.
Unlike their Srivijayan contemporaries in far-off Sumatra, the royals of Mataram were not, initially, concerned with maritime trade. There were no safe anchorages on the southern shores of Java, and the ports of the north coast were some distance away. What was more, they had no particular impetus for internationalism at this early stage: the region produced all they could possibly need. Rice and pulses grew in the irrigated fields, and the forests turned out fruits and seeds and dyestuffs. There were mines for salt, fish in the rivers, and a solid tradition of craftwork. Much of the excess produce did, eventually, find its way out to the ports, but it travelled by way of middlemen in the markets that rotated through rural settlements on a five-day cycle, and the rulers probably had little direct involvement in the trade.
The power of these Javanese rulers was rooted in water of the fresh, rather than salt, variety. Any given tranche of farmland had to be irrigated to keep its soils turning out crops through the fiery dry months in the middle of the year. Complex networks of dykes and ditches criss-crossed the countryside, and keeping them all free from leaks demanded cooperation across any number of banyan-shaded villages. The councils and collectives that managed these water networks were the earliest forms of organised government, and from time to time the chief of a particular network would decide to extend his influence beyond the head of his longest irrigation trench.
Villages became small kingdoms, and something not unlike the Srivijayan hub-and-spoke model of power developed in Central Java. A king would have his own timber-built palace, known as a kraton, with a clutch of directly ruled fields and hamlets close by. Ranged at a distance around this fulcrum were the seats of other feudal lords, vassals of the central chief but each with its own spiral of subject villages. Still further afield would be other fiefdoms owing only the most notional allegiance to the centre. Across the network the rudiments of a tax system were already in place. Each village owed a certain time-honoured portion of their crops to the overlord, and each community owed a tranche of their time too, a fixed amount of man-hours to be given over to labouring on the landlord’s behalf. These obligations in time and kind might be owed only to the local village chief, or they might be passed all the way up the chain of command to the king himself.
The whole set-up allowed separate dynasties to travel side-by-side in the same region, vying for ascendency without annihilating each other. Over the generations the hub of the power-wheel might shift to some new village as a different dynasty won supremacy; the spokes leading to outlying vassals would realign, but the power structures and systems of tax and tribute would remain essentially unchanged. In Mataram there were two of these jostling lineages, twin clans which would leave the most monumental of marks on the Javanese landscape.
