A Thousand Years Before Melaka

In 1834, a British army captain named James Low picked up a slab of stone in Province Wellesley.

There was writing on it. Not Jawi, not roman letters, not Chinese characters. It was Sanskrit, in the Pallava script of southern India.

When it was finally read, the stone turned out to be neither a royal decree nor a declaration of war. It was a sailor's prayer. A ship's captain named Buddhagupta, asking that his voyage be a safe one.^1,2

It is roughly fifteen hundred years old.

Stop on that for a moment. A ship's captain, far from home, stopping on this coast and cutting a prayer into stone so that he might reach his destination alive.

That stone is not a monument. It is a receipt — proof that this land was a stop on the world's sea roads a very long time ago.

The first kingdom with a name

The earliest kingdom on this land whose name survives in a document was not written down by anyone here. It was written down by court clerks in China.

The records of the Liang dynasty mention a kingdom called Lang-ya-xiu — the one we know today as Langkasuka. It sent its first mission to China in 515 CE.^3,4

Five hundred and fifteen. Nearly nine hundred years before Melaka.

And here a small necessary honesty, because this gets misstated often: although a Malay literary chronicle places Langkasuka on the west coast as Kedah's predecessor, nearly every other source — Chinese, Arab, and the archaeology — places it on the east coast, near Pattani in what is now southern Thailand.^4,3

Modern national borders were drawn in the twentieth century. The old kingdoms knew nothing about those lines.

Dozens of temples in one valley

Walk into the Bujang Valley in Kedah today and you will walk among dozens of candi — the ruins of Hindu and Buddhist religious sites scattered the length of the valley.^5,6

The first systematic excavation was carried out by a pair of British researchers, H.G. and Dorothy Quaritch Wales, in the 1930s.^5,7

Among the things they found: gold foil stamped with the image of Nandi — the bull that is Shiva's mount. A trident. Nine-chambered stone reliquaries. A bronze Buddha.

On that same ground, today, the call to prayer sounds five times a day.

Both are true. Both are our history.

The hardest admission in this chapter

Now the uncomfortable part — and we are writing it precisely because it is uncomfortable.

You may have heard that Sungai Batu in the Bujang Valley is the oldest civilisation site in Southeast Asia, older than Angkor Wat and Borobudur, with dates reaching back to 788 BCE. That figure has been in newspapers, in speeches, and in our own conversations, carried with pride.

Here is the problem.

Those earliest dates rest on a handful of single radiocarbon samples at the very early end of a wide range. In 2021, two researchers applied Bayesian chronological modelling to the site's whole radiocarbon dataset — a method that weighs all the samples together rather than picking the most extreme one. The result places the iron-smelting activity at Sungai Batu between the 2nd and 10th centuries CE. Not BCE.^6,8

Several international archaeologists have also raised doubts about the original dating method. And the Department of National Heritage itself officially accepts the 2nd-to-5th-century CE range, while stating that it remains open to debate.^8,6

This does not mean Sungai Batu is unimportant. Iron smelting in 2nd-century Kedah, inside a trade network reaching halfway around the world, is a serious technological achievement on its own terms.

But it is not 788 BCE. And we do not need it to be 788 BCE.

A temple flattened in 2013

Now the part that hurts.

In 2013, a candi at Sungai Batu — known as Candi 11 — was flattened by a developer's excavators. It was done without announcement. The public only learned of it when a researcher, Datuk V. Nadarajan, came across it and filed a police report.^9,10

How could this happen? Because the site was not gazetted as national heritage, and the land was privately owned. Legally, there was nothing to stop it.

The structure that was flattened was in fact a 1970s reconstruction on the original site. But what was lost was not the bricks — what was lost was the soil layers beneath them, the arrangement of the artefacts, and all the information that can only be read while the ground is still undisturbed.

That cannot be replaced. Not with money, not with an apology, not with anything.

We argue with great energy about whether the site is 788 BCE or 2nd century CE. Meanwhile, part of the very site being argued over has already been shovelled into a lorry.

Kedah's name, cut into stone in India

In 1025 CE, a Chola fleet out of southern India launched a naval raid across the Straits of Melaka.

We know this because it was carved into stone — not here, but in a temple complex at Thanjavur in India. The Tamil inscription of Rajendra Chola I records the capture of Sangrama-vijayottunga-varman, king of Kadaram. Kadaram is Kedah.^11,12

Picture that: the name of a Malaysian state cut into a temple wall in India a thousand years ago, as the record of someone else's victory.

It has to be stated precisely: historians generally read this as a raid to seize tribute and assert Chola naval power — not a permanent conquest. Kedah did not become a Chola province.^12,11

Srivijaya — and how much we actually know

The name that comes up most often in this stretch of history is Srivijaya — a maritime power centred on Palembang in Sumatra.

But honesty is needed again here, because Srivijaya is often told as though it were an empire with borders and governors.

Modern archaeology confirms its centre at Palembang, but characterises Srivijaya not as a bureaucratic empire — more accurately as a pliable mandala: a network of ports and shifting loyalties, held together by trade and recognition rather than by direct administration. Historians working from the textual sources describe its influence as symbolic hegemony maintained through tribute, rather than day-to-day rule.^13,14

So the next time you read "Srivijaya ruled Kedah" — slow down. The relationship was real, but its shape is still argued over by people who have studied it their whole lives.

Then, a different kind of stone

And at some point, the kind of inscription being found on this land changed.

At Kuala Berang in Terengganu, a stone was found whose writing was no longer Sanskrit. It was Jawi — the Arabic alphabet adapted for the Malay language. Nor was the content a sailor's prayer: it was a ruler's proclamation, calling on his subjects to uphold Islam, and setting out ten provisions of law.

Its date is read as 702 AH — corresponding to 22 February 1303 CE. It is now on the UNESCO Memory of the World Register as the earliest evidence of Jawi writing in the Malay world.^15,16

From a ship captain's Sanskrit prayer to Islamic law written in Jawi. Eight hundred years between two stones.

But this story is not finished.

Because towards the end of the 1300s, a man who had lost a war came ashore at a river mouth on the west coast. The Chinese court records call him by one name. The Malay court chronicle calls him by another. And the title he carried — Parameswara — comes from Sanskrit, the language of the very stones we have just been reading.

Who he actually was, and what religion he held, is one of the hardest questions in the whole of our history.