Srivijaya

Power over a passage

Drafted from Hall, A History of Early Southeast Asia: Maritime Trade and Societal Development, 100-1500 (2011)formation

Srivijaya held no territory worth the name and ran the trade of half of Asia for six hundred years.

It sat on the Melaka and Sunda straits — the only sea route between the Indian Ocean and the South China Sea — and made its living by making that passage safe for those who called at its ports and unsafe for those who did not. Chinese sources describe it, for six centuries, simply as the place the southern trade came through.

How you hold a strait

You cannot garrison a coastline that long, and Srivijaya never tried. What it did instead was buy the loyalty of the people who would otherwise have robbed the ships.

The coastal chiefs and the crews of the river mouths — the orang laut, the sea peoples — were bound to the ruler by a combination of redistributed trade wealth, ritual authority, and oaths sworn over inscribed stones that threaten spectacular curses against the disloyal. The arrangement was that they raided for Srivijaya rather than against it. A polity of this kind has no army in any ordinary sense; it has a protection agreement with the pirates, renewed continuously.

That is a genuinely different kind of state from the ones on either side of it. A kingdom on a floodplain counts its people and taxes its harvests, and its power ends where the fields end. This one had no land tax, no standing force and no frontier. Its whole asset was a reputation for making a particular stretch of water predictable, and every institution it had existed to maintain that reputation.

Which is why it cannot be measured the usual way

There is no figure for how large Srivijaya was, and there is not going to be one.

Square kilometres of land under state control is the standard measure of a premodern polity's size, and it describes this one about as well as acreage describes a shipping line. Taagepera, whose tabulations supply extent figures for most of the empires of the period, does not tabulate it. Drawing an area off a map of the straits would produce a number, and the number would be a picture of an idea rather than a measurement of anything.

So the honest answer is that Srivijaya was very large and unmeasurable, and the things worth counting about it are the ports that paid, the monks who stayed, and the centuries it lasted.

Siddhayatra

Drafted from Hall, A History of Early Southeast Asia: Maritime Trade and Societal Development, 100-1500 (2011)expansion

The Kedukan Bukit inscription, cut on a small stone near Palembang and dated to 683, records that a ruler called Dapunta Hyang set out on a siddhayatra — a journey for magical power or success — with twenty thousand troops, took a boat, and afterwards Srivijaya was prosperous.

It is the oldest dated text in Malay, in any form, anywhere.

The language

That fact is worth stopping on before the politics.

These inscriptions are written in Old Malay, in a script derived from Pallava south India, and Malay was not the language of Palembang's hinterland or of most of the places Srivijaya dealt with. It was the language of the coasts and the ports — the trade language of the straits — and a polity that chose to legislate and to curse in it was making a decision about who it was addressing.

That decision outlasted everything else here. Malay went on being the lingua franca of the archipelago for a thousand years after Srivijaya stopped existing, was adopted by the Dutch administration for the same reason the traders had adopted it, and is now the national language of Indonesia and Malaysia under two names.

The letters themselves were borrowed, arriving from south India with Buddhism and Sanskrit. Srivijaya invented no writing system. What it did was put a trade vernacular into one, which is the smaller act and the more consequential.

What the expedition was for

The inscriptions of the 680s are a set: Kedukan Bukit, Talang Tuwo, Kota Kapur, Karang Berahi and Palas Pasemah, spread from Palembang up the coast to Bangka island and into Jambi.

Read together they describe an expansion. The outlying ones stand at exactly the places a polity would need if it wanted the Melaka strait and the Sunda strait both, and they carry the loyalty oath the fourth chapter is about. Kota Kapur, on Bangka, adds a line about an expedition against a land that would not submit.

So the founding was not a conquest of territory. It was a set of stones placed at river mouths, each carrying a curse, each addressed to whoever controlled the boats there.

The Sailendras, and Borobudur

In the ninth century Srivijaya's dynastic history entangles with Java's in a way nobody has fully untangled.

Balaputradewa, the king under whom Srivijaya was at its height, appears in a copper-plate charter from Nalanda in Bengal, where he endows a monastery and is described as a Sailendra — the dynasty that built Borobudur on Java, the largest Buddhist monument in the world.

Whether the Sailendras were a Javanese house that took over Srivijaya, a Srivijayan house that ruled Java for a period, or two branches of one family is argued and not settled. What the Nalanda plate establishes is that a king at Palembang was funding a Buddhist university in northeast India, and that an Indian king recorded it on copper.

A Sumatran ruler's endowment in Bengal, a Javanese monument, and a dynasty claimed by both islands: it is the ordinary condition of the evidence here that Srivijaya is visible in everybody's sources except its own.

A thousand monks

Drafted from Hall, A History of Early Southeast Asia: Maritime Trade and Societal Development, 100-1500 (2011)peak

Yijing, a Chinese Buddhist monk, sailed for India in 671 and stopped at Srivijaya on the way. He stayed about six months to learn Sanskrit grammar, went on to Nalanda for a decade, came back, and stayed at Palembang again — for years this time — writing and translating.

His advice to Chinese monks heading for India was to stop here first and do a year or two of preparation, because the Sanskrit was as good and the discipline the same. He describes more than a thousand monks in residence.

Traffic in both directions

Srivijaya did not invent Buddhism and converted nobody at spearpoint. What it did was be the place between China and India where the texts, the language and the discipline were available, for six hundred years, on the route everybody had to take anyway.

And the traffic ran both ways. Yijing came south to prepare for India. Atisha — the eleventh-century master whose teaching reshaped Tibetan Buddhism, and whose lineage the Tibetan schools were still arguing over centuries later — is recorded as having studied for twelve years under a teacher at Srivijaya before going north.

A Bengali monk trains in Sumatra and reforms Buddhism in the Himalayas. That is the kind of thing a polity on a trade route does that has nothing to do with trade.

What the monasteries were for

The connection between the monks and the tolls is not incidental.

A state with no land tax and no army has exactly two things to spend its customs revenue on. One is keeping the crews loyal. The other is building the reputation that makes ships want to call in the first place — and a ruler who endowed monasteries, hosted foreign scholars and funded a monastery at Nalanda in Bengal was not being pious at the expense of policy.

He was making Palembang a place worth stopping at. A shipmaster choosing between harbours is choosing on safety, on price, and on whether the place is consequential enough to be worth the detour. Srivijaya spent six centuries ensuring the answer to the third question was yes, and the monks were a substantial part of how.

Curses in stone

Drafted from Hall, A History of Early Southeast Asia: Maritime Trade and Societal Development, 100-1500 (2011)aside?What aside means

Srivijaya left few monuments and no chronicle. What it left is a small set of seventh-century inscriptions in Old Malay, and they are unusually direct about how the polity worked.

The Telaga Batu stone carries an oath. It lists the people bound by it — princes, commanders, judges, scribes, naval captains, merchants — and specifies what happens to them if they conspire against the ruler: an elaborate curse, invoked over water poured across the stone and drunk by those swearing it. Those who stay loyal are promised prosperity.

A state describing its own weakness

That is the political technology of a thalassocracy set out in its own words, and what is striking is how frankly it admits the problem.

A ruler who commanded his subordinates would not need to curse them. The oath is addressed precisely to the people who could defect and could not be prevented from defecting — the captains and the coastal chiefs whose loyalty was the whole apparatus of the state. Water, a stone and a supernatural penalty are what you reach for when you have no garrison to send.

It also tells you what the ruler's authority was understood to be. Not a right of command backed by force, but a power that could reach a man who broke his word, anywhere, including at sea and out of sight. For a polity whose subordinates spent most of the year beyond anybody's reach, a god who could follow them was not a decoration on the constitution. It was the constitution.

What survives, against what it ran

Set the evidence beside the achievement.

A handful of stones near Palembang. A Chinese pilgrim's travel notes. Tribute records kept by other people's courts, in other people's languages, for their own purposes.

From that: six hundred years, the trade of half of Asia, and a monastic centre that Bengali and Chinese scholars crossed oceans to study at. Srivijaya is the clearest demonstration in the region that the quantity of surviving evidence has almost nothing to do with the importance of what produced it.

1025

Drafted from (eds), Nagapattinam to Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia (2009)contraction

Rajendra I of the Chola empire sent a fleet across the Bay of Bengal and attacked the ports of the Melaka strait. The great inscription at Tanjavur lists what was taken: something like fourteen places by name, Srivijaya's capital among them, and it claims the king was captured along with a gate of jewelled splendour.

The Cholas did not stay. There is no Chola administration in Sumatra, no garrison, no tribute record of any length. They came, took the ports, and sailed home.

Why a raid mattered more than a conquest would have

Because what Srivijaya sold was not territory. It was a guarantee.

Ships paid to call because calling was safe and predictable. The crews who could have been pirates were bound not to be. The ruler's authority was understood as a power that reached anyone who broke his word. All three of those things — the tolls, the oaths, the curse — depend on Srivijaya being the arrangement that made the strait work.

A fleet that crosses an ocean, takes the capital and leaves has demonstrated in a single season that the arrangement does not hold. Not to Srivijaya, which recovered enough to go on sending embassies to China for another century, but to every port that had been accepting its terms and every shipmaster deciding where to stop.

After 1025 the Chinese records begin listing the Sumatran ports separately rather than as dependencies of one power. That is the whole of what the raid did, and it was sufficient.

Why they came

Argued about. The candidates are a trade dispute, with Srivijaya obstructing Chola merchants' direct access to China; a response to Srivijayan interference with Chola embassies; a bid to open the strait for the Tamil merchant guilds then operating across the Bay of Bengal; and plunder, which royal inscriptions never give as a motive and which is usually part of one.

What is not in doubt is that relations before it had been good. Srivijaya endowed a Buddhist temple at Nagapattinam on the Chola coast in about 1006 with Chola royal consent, and the charter recording it survives. Nineteen years later the same two courts were sacking each other's ports.

Who the attackers were

The Cholas are worth knowing about, because a raid like this is otherwise hard to credit.

They were a south Indian dynasty who held the Coromandel coast for roughly three centuries and built the largest navy in the Indian Ocean. Their temple architecture at Tanjavur defines a tradition; their bronze casters produced the Shiva Nataraja figures that are among the most reproduced sculptures in the world; their administration is documented in tens of thousands of temple inscriptions, which is a denser record than almost anything in Southeast Asia from the same centuries.

They were also, unusually for an Indian dynasty of the period, willing to project force across open sea — Sri Lanka, the Maldives, and in 1025 the far side of the Bay of Bengal. Srivijaya's misfortune was to be the richest thing within reach of the one Indian power that had built a fleet capable of getting there.

The position outlives the polity

Drafted from Hall, A History of Early Southeast Asia: Maritime Trade and Societal Development, 100-1500 (2011)end

Srivijaya did not fall so much as stop being the name for the thing.

The first move was internal. In 1079 the centre shifts from Palembang up the coast to Jambi-Melayu, which had been a rival and then a dependency. Chinese records begin naming the Melayu port where they had named the Palembang one. That is the clearest sign of what the Chola raid had actually done: the strait still had to be passed and somebody still had to be paid, but the somebody had changed.

For two centuries after that there is very little. No ruler of the polity is named in anything after the eleventh century. The ports go on trading, the tolls go on being collected, and nobody records who is collecting them.

1275

Kertanegara of Singhasari, on Java, sent an expedition across to Sumatra — the Pamalayu — and afterwards the Melayu polity that had been carrying Srivijaya's position stops appearing as an independent power.

Nobody annexed anything. There was no Javanese administration in Sumatra, no tribute list, no garrison; Singhasari itself was overthrown at home seventeen years later. What ended was simply the last claim that the straits were run from Sumatra by a Sumatran power.

Eighteen years after the Pamalayu, Majapahit was founded on Java, and for the next three centuries the centre of gravity of the archipelago sat there. The thing Srivijaya had been — a single authority selling safe passage through the straits — was not reassembled in Sumatra. When it reappeared it was on the other side of the water, at Melaka, and the man who built it came out of Palembang.

A kingdom nobody remembered

Drafted from Cowing, The Indianized States of Southeast Asia (1968)afterlife

By 1900 European scholarship had catalogued the temples of Java and the inscriptions of Cambodia, and had no name at all for a six-hundred-year maritime power in between.

It had the evidence. It just had it in three unconnected piles: Chinese records of a kingdom called Shilifoshi that the southern trade came through; Arabic and Indian references to a place called Zabag or Sriwijaya; and Old Malay inscriptions from around Palembang that nobody had linked to either.

In 1918 George Cœdès, a French epigrapher, put the three together and showed they were one polity. Paul Wheatley later called it "possibly the most significant contribution ever made to the progress of Southeast Asian history" — a large claim for a short paper, and not obviously an overstatement, given that it produced an empire.

What had actually been forgotten

The name, and only the name. It is worth being exact about this, because "a kingdom nobody remembered" is the sort of phrase that runs away with itself.

Palembang was never forgotten. Malay court tradition went on routing its dynasties back to it for centuries — and the most consequential of those genealogies is the one that produced Melaka.

The founder of Melaka was a prince called Parameswara who was, on Albuquerque's account, a native of Palembang, and who was married to a princess of Majapahit. He revolted, probably after the death of Hayam Wuruk in 1389, and took refuge at Tumasik — modern Singapore, and of all the peninsular possessions of the kingdom of Palembang the one furthest from Ayutthaya's reach. He killed its chief, reigned there a few years, was driven out by Ayutthaya, fled to Muar and then Bertam and finally to Melaka, where a Chinese eunuch envoy found him established in 1403.

The Sejarah Melayu, the Malay Annals, traces his line back through Palembang to the kings of Srivijaya, and then onward to Alexander the Great. It was composed nearly six centuries after the last Srivijayan ruler anyone can name, by a court writing in exile after its own city had fallen — so it is a claim rather than a record, and the second half of it is plainly a claim.

But the first half is a memory of something real. A man out of Palembang founded a port on the other side of the strait that did exactly what Srivijaya had done — sold safe passage, bound the sea peoples, took the tolls — and his descendants said so, in those terms, on the ground that they came from the place that had done it first.

What the rediscovery gave back

So the sequence is stranger than either "forgotten" or "remembered" allows.

Srivijaya's institutions were carried on at Jambi, and its position was rebuilt at Melaka by a family who knew where they had come from. What was lost was the word: the polity's own name for itself, which survived in Chinese transcriptions, Arabic travel accounts and Sanskrit-Malay inscriptions that nobody had thought to read alongside each other.

For six hundred and forty-three years the largest maritime power in the history of the region existed in the record under three different names in three different languages, and belonged to nobody. Cœdès gave it back its own.