The Kingdom of Kediri

The division

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

In 1042 Airlangga, who had spent thirty years putting eastern Java back together after it fell apart, divided the kingdom he had rebuilt between his two sons and retired to live as an ascetic.

The Javanese account gives the reason as a father's even-handedness. The result was two kingdoms, Janggala and Panjalu, which fought each other; Panjalu won, took the name of its capital's district, and became Kediri.

Java's dynastic chain runs from Kediri to Singhasari to Majapahit, and Kediri is where it starts.

Cowing, The Indianized States of Southeast Asia (1968) · Government of Nganjuk Regency · Public domain

Nobody knows who founded it

Airlangga is the obvious candidate for founder and he is the wrong one. He ruled Kahuripan, the kingdom that was divided; Kediri is one of the halves, and he did not rule it.

The first ruler of Panjalu is named in the inscriptions, and there is nothing attached to the name. No reign anybody can date, no act, no source that puts him in a sentence. He is a name on a stone, and that is the whole of him.

That is a more common situation than the tidiness of most dynastic lists suggests. Srivijaya, Aksum and Champa are all in it — kingdoms with a documented history and an undocumented beginning — and in each case the temptation is to promote the most plausible candidate into the empty slot and move on. A name supplied that way reads exactly like a name somebody wrote down at the time, which is the whole reason for leaving it visibly empty.

The Bharatayuddha

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

Jayabhaya reigned at Kediri from about 1135 to 1157, and what makes his reign the kingdom's height is not a battle or a boundary.

It is the Bharatayuddha, an Old Javanese kakawin composed at his court in 1157 by the poets Mpu Sedah and Mpu Panuluh, which takes the war of the Mahabharata and makes it Javanese. It sits at the head of a body of court literature — the Smaradahana, the Kresnayana, the Hariwangsa — that this kingdom produced and that everything written in Java for the next seven hundred years was measured against. Majapahit's Nagarakertagama, composed two centuries later, is in the same tradition and the same metres.

The Bharatayuddha is still performed. The wayang repertoire of Java and Bali is built on it.

Cowing, The Indianized States of Southeast Asia (1968) · Eka343 · CC BY-SA 4.0

What a literature is worth

Kediri conquered little, converted nobody, and left no successor claiming to be it. Singhasari took its territory by force in 1222 and never pretended to be continuing anything.

By any measure of power it is a middling kingdom on a Javanese river, and it is remembered because of what was written at its court.

That is a genuinely awkward kind of achievement to weigh. A conquest has a date and a map. A conversion has a boundary it crossed. A literature has neither: it is not a thing that happened in a year, it did not travel by an identifiable route, and its influence is visible only in the fact that everybody afterwards kept doing it.

But the Bharatayuddha is still performed. The wayang repertoire of Java and Bali is built on it, and Javanese poets were still composing in its metres in the nineteenth century. Seven hundred years of a literary tradition came out of one court in the middle of the twelfth century, and no account of what Kediri was for can leave it out.

Cowing, The Indianized States of Southeast Asia (1968) · Tropenmuseum · CC BY-SA 3.0

What the inscriptions are

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

Almost everything datable about this kingdom comes from copper plates and stones issued by its kings, and they are nearly all the same kind of document: a grant of sima status.

A sima is land whose revenue is assigned away from the crown — to a temple, to a religious community, sometimes to a village in return for a service performed. The charter records who granted it, which land, what the boundaries are, what obligations are lifted, what the ceremony cost, who witnessed it, and a curse on anyone who interferes.

They are, functionally, the state's tax records written from the receiving end.

Hall, A History of Early Southeast Asia: Maritime Trade and Societal Development, 100-1500 (2011) · Isidore van Kinsbergen / Leiden University Libraries · Public domain

Medang was governed the same way three centuries earlier and is known from the same class of document.

What that does to what can be known

Almost everything anyone says about how Kediri worked is an inference from that one genre.

We can say the state lived on land, because the charters exempt land from obligations that must therefore have existed and are listed by name. We can say it raised labour and service through territorial officials, because those are among the obligations being lifted. Neither is stated anywhere as a rule; both are read off the back of a tax exemption.

How the throne passed is simply not answerable, and one reason among several is that charters do not discuss succession. They give a reigning king's titles and his ancestry as he wished them stated, and a king who took the throne irregularly does not say so in the preamble to a land grant.

What a civilisation chose to carve

The general point is worth having, because it applies wherever the evidence is epigraphic.

What survives is whatever that culture happened to inscribe on durable material, and the genres differ enormously. Angkor carved temple foundations, so we know how many people were assigned to a particular shrine and what they were fed. Assyria carved royal annals, so its history reads as an unbroken sequence of campaigns. Kediri carved tax exemptions, so it comes down to us as a fiscal system with a literature attached and no politics in between.

None of those is a description of what the kingdom was. Each is a description of what it wrote on stone.

Ganter

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

In 1222 Ken Arok, a man of obscure origins in the service of the lord of Tumapel, defeated and killed Kertajaya at Ganter. Kediri had held eastern Java for a hundred and eighty years, and stopped.

Hall, A History of Early Southeast Asia: Maritime Trade and Societal Development, 100-1500 (2011) · Michael Gunther · CC BY-SA 4.0

Ken Arok's own story, as the Pararaton tells it, is one of the strangest in Javanese literature: a man of obscure and possibly divine parentage, a cursed dagger commissioned from a smith whom he then murdered with it, and four generations of revenge killings running down from that weapon through the dynasty he founded.

Ganter is where a hundred and eighty years of Kediri end and that story begins.

And then it came back

Seventy years later a lord of Kediri named Jayakatwang killed Singhasari's last king and restored the kingdom that had ended here.

It lasted about a year. Kublai Khan's expedition to Java arrived in 1293 looking for Singhasari's king, found Jayakatwang on the throne instead, and was manoeuvred by Raden Wijaya into destroying him — after which Wijaya turned on the Mongols, drove them out, and founded Majapahit.

So Kediri ends twice, seventy-one years apart, and the second ending is the more consequential: it is the accident that produced the largest state Java ever had.

Jayabhaya's prophecy

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

Jayabhaya reigned at Daha in the middle of the twelfth century and commissioned the Bharatayuddha. That is the historical claim.

The other thing attached to his name is not historical at all, and it has had by far the longer career.

The Jangka Jayabaya is a body of Javanese prophetic literature, attributed to him, that circulated in manuscript for centuries and predicted the course of Javanese history: a sequence of ages, foreign rule, and a Ratu Adil — a just king — who would come and set things right. Its actual composition is much later than the twelfth century and the texts we have are eighteenth- and nineteenth-century, but the attribution is to this king, and the attribution is the point.

Where it turns up

In the nineteenth century, in Javanese risings against the Dutch, where a leader identifying himself against the prophecy is a recurring pattern. In 1942, when one of the verses — read as predicting that Javanese rule would return after a period of yellow men lasting the life of a maize plant — circulated widely as the Japanese arrived. It is still quoted.

Jakarta, Indonesia · Cowing, The Indianized States of Southeast Asia (1968) · Raden Saleh · Public domain

That matters more than it sounds. Javanese rebellion, from the seventeenth century onward, was rarely framed as a bid for a throne. It was framed as a restoration — an age turning, a just king arriving, the order coming back — and the body of text that made it legible in those terms carries Jayabhaya's name on it.

Nobody claimed to be Kediri, in the end

No later Javanese dynasty of any permanence asserted that it was continuing this one. A lord of Kediri did briefly seize Singhasari's throne in 1292 in the name of the kingdom Ganter had ended seventy years before, but his own restoration lasted about a year before Majapahit's founder destroyed it in turn; Majapahit traced its own legitimacy to Singhasari, and the sultanates after it to Majapahit. The one revival Kediri's name did produce burned out almost as soon as it succeeded, and no dynasty after it looked back further than Singhasari.

What Kediri kept instead of an heir was a name that would not stop working. Nearly eight centuries after Jayabhaya died, in 1942, Javanese were reading a verse attributed to him about a period of yellow men lasting the life of a maize plant, and drawing conclusions about the Japanese.

A kingdom nobody claimed to inherit, whose king's name was still being used to interpret an invasion in the twentieth century. Political succession and being remembered turn out to be almost unrelated.