Srivijaya and Majapahit: Maritime Powers of the Indonesian Archipelago

Srivijaya and Majapahit are the two names that stand in for a thousand years of archipelago history, and they are usually introduced with confident maps showing large coloured territories. Both states were real and important. The maps are the least reliable thing about them, because these were powers built on controlling movement between islands rather than on administering continuous ground — and because the two pose almost opposite problems of evidence.

Srivijaya is known mostly from outside: foreign records, a scattering of inscriptions, and later references, with a physical footprint that archaeologists have argued is thinner and later than the texts imply. Majapahit has a located capital, a substantial contemporary Javanese account, and a site so damaged before modern excavation that basic questions about its layout stayed open for a century.

Srivijaya: the state that is easier to cite than to excavate

The standard account places Srivijaya's centre at or near Palembang in southern Sumatra, controlling the straits through which shipping between the Indian Ocean and the South China Sea had to pass.

The complication came from the ground. Archaeological work on Palembang published in Asian Perspectives argued for the lateness of early cities in southern Southeast Asia: the urban remains there are later and slighter than a great entrepôt implied by the textual record would lead you to expect. That finding did not make Srivijaya fictional. It reframed the question — from "where is the city?" to "what kind of polity leaves this little?"

Plausible answers include a network of riverine and coastal settlements rather than a single capital; a court that was mobile; building in perishable materials; and a power whose wealth sat in shipping and transhipment rather than in monumental construction. Recent scholarship on the eastern Indian Ocean between about 1000 and 1500 has moved in this direction generally, reading epigraphic, archaeological, art-historical, literary, textual and shipwreck evidence together and giving merchant diasporas as much agency as rulers.

That is a different mental model from an empire, and it is closer to how the overland networks worked too, as the Silk Road for beginners sets out: a system of nodes and intermediaries rather than a territory with a border.

The monuments belong to somebody else

One correction is worth making early, because the error is extremely common.

UNESCO dates Borobudur to the eighth and ninth centuries in central Java, describing a monument built in three tiers — a pyramidal base with five concentric square terraces, a conical trunk with three circular platforms, and a monumental stupa at the top — with low reliefs covering some 2,500 square metres and seventy-two openwork stupas around the circular platforms, each containing a statue of the Buddha. UNESCO dates the Prambanan compound to the tenth century, calling it the largest temple compound dedicated to Shiva in Indonesia, with three central temples for Shiva, Vishnu and Brahma carrying reliefs of the Ramayana.

These are central Javanese monuments produced by Javanese dynasties. They are not Srivijayan, and they are not Majapahit either — they predate Majapahit by three to five centuries. Assigning them to whichever archipelago name a reader already knows is exactly the compression this article exists to prevent.

Majapahit: a located capital and a damaged one

Majapahit held power in Java from the thirteenth to the sixteenth century, and its royal city has long been known to have stood at Trowulan, about fifty kilometres south-west of Surabaya.

Knowing roughly where a capital was and being able to map it are different things. The scholarship on Trowulan describes what remains of its former glory as isolated piles of broken bricks in an archaeologically mutilated landscape littered with craters dug by colonial-era asset strippers and treasure hunters. The specific puzzle that occupied Dutch and later archaeologists was the position of the fourteenth-century kraton, the rulers' palace, which had to be reconstructed from fragmentary survey records and textual description.

The textual side is unusually good. In 1365 the Javanese author Prapanca produced a description of the Majapahit court — a contemporary account written from inside the kingdom, which is rare for Southeast Asian states of this period and which has been used, with care, to try to fit the physical remains to a described plan.

Reading a court poem as a site plan is a demanding exercise, and it is a clean illustration of the distinction drawn in primary and secondary sources: a contemporary text is close to events without being neutral about them, and a royal panegyric describes an ideal arrangement as readily as an actual one.