Belitung Shipwreck — The Arab Dhow That Proved the Maritime Silk Road
The Belitung shipwreck is an Arab dhow that sank around 830 AD in about 17 meters of water off Belitung Island, Indonesia, on its way back from China with roughly 60,000 Tang-dynasty ceramics, gold and silver. Local fishermen found it in 1998, and it is described as the first ancient ship of its Arabian type ever found and excavated.
The wreck is famous for two reasons. Its cargo is the largest single collection of Tang dynasty artefacts found in one place outside China, and its excavation by a commercial company, rather than a state archaeology team, made it one of the most argued-over finds in underwater heritage.
How was the Belitung wreck found and excavated?
Local fishermen in the Gelasa Strait came across the site in 1998, at a depth of about 17 m and roughly 1.6 km off Belitung Island. Word spread quickly. The location was bought from the fishermen, an Indonesian salvage company obtained the excavation licence, and that company brought in Seabed Explorations, a company led by Tilman Walterfang, to finance and carry out the work under a cooperation agreement. The Indonesian Navy provided security, paid for by Seabed Explorations, because looters were already working the site.
The recovery ran in two campaigns. The first began in August 1998 and the second followed in 1999, with Seabed Explorations paying for naval patrols to guard the site through the monsoon in between. Some looting took place between the two. The maritime archaeologist Michael Flecker directed the excavation in the second season, recorded the surviving hull and later published the wreck in scholarly journals and exhibition catalogues. Walterfang described the overall situation as “less than ideal” and said the Indonesian government, fearing looters, had ordered a round-the-clock recovery.
The ship survived because it was buried. A layer of sediment covered the lower hull and sealed it off from shipworms, which normally destroy exposed wood in tropical water within a few years. That is why so much of a ninth-century hull was left to record.

What kind of ship was the Belitung dhow?
The vessel was about 18 m long and 6.4 m wide, built on a keel some 15.3 m long and 14 to 15 cm thick that seems to have survived intact. The bow raked forward at about 61 degrees, and the stem post was joined to the keel with mortise-and-tenon joints and lashed with rope 16 mm thick. Its planks were not nailed or pegged. They were sewn to each other and to the frames with thin coconut-fibre cord passed through holes drilled every 5 to 6 cm, and a keelson resting on half-frames stiffened the bottom.
Sewn boats were not new. Flecker pointed out that fully stitched hulls were built on the African coast, in Oman, in the Red Sea, in India and as far as the Maldives, and that the Byzantine historian Procopius described ships of the “Indian Seas” whose planks were sewn together in the sixth century. A late-Tang Chinese text, the Ling biao lu yi, likewise described foreign merchant ships stitched with the fibre of coir palms. What was missing until 1998 was a surviving example of such a ship from this period, still with its cargo. Belitung is described as the first ancient ship of this Arabian type ever found and excavated.
Was it really an Arab ship? What the timber says
The hull’s origin has been argued over since the excavation, and the answer depends on how you weigh the evidence. Wood samples were sent to Australia’s CSIRO, where the wood specialist Jugo Ilic identified what he could; many samples had lost too much cellulose to identify. Teak was used for the through-beams, a cypress, possibly Cupressus torulosa, for the ceiling planks, a rosewood of the genus Dalbergia or Pterocarpus for the stem post, and the breadfruit genus Artocarpus for a wooden box found in the stern. The hull planks were probably Amoora, and one frame was probably Amoora or Afzelia, a genus whose main species grow only in a band across central Africa.
From the hull form, timbers and methods, Ilic concluded that the ship was of Indian or Arabian origin, with India the more likely place of construction; Arabia could not be ruled out, because timber of these kinds was imported into the Gulf for shipbuilding. Flecker went further. In a 2010 catalogue chapter he wrote that the Belitung wreck “is an Arab vessel.” He also noted resemblances to the lashed-lug boats of Austronesian Southeast Asia, which used a different way of fastening planks and frames. Describing the ship as built of African hardwood, as many summaries do, oversimplifies the evidence: only one frame may be African wood, and most of the species identified are Asian.
Belitung wreck timeline
| Date | Event |
|---|---|
| 826 AD | Date inscribed on one Changsha bowl |
| c. 830 AD | Dhow sinks off Belitung on its return voyage |
| 1998 | Local fishermen find the wreck |
| 1998–1999 | Two-phase excavation by Seabed Explorations |
| 2005 | Singapore buys the cargo for about US$32 million |
| 2011 | Exhibition opens in Singapore; US premiere postponed |
| 2008–2010 | Replica Jewel of Muscat built in Oman and sailed to Singapore |
| 2015 | Permanent display announced at the Asian Civilisations Museum |
| 2023 | Natali Pearson publishes Belitung: The Afterlives of a Shipwreck |
Why was it so far off course?
The usual route home from China to the Gulf ran south through the South China Sea, past southern Vietnam, through the Singapore Strait and up the Strait of Malacca between the Malay peninsula and Sumatra. Belitung lies about 610 km south-east of the Singapore Strait, closer to the route between the Java Sea and Malacca. Nobody knows why the dhow was there. It may have been heading for a port in Java or Sumatra to trade before crossing the Indian Ocean, it may have been blown off course, or its crew may have been using a route that left no other record.
The wider setting helps explain why such a ship was in these waters at all. The kilns at Changsha lay roughly 3,500 km north-north-east across the South China Sea, and the cargo had to be carried down rivers to the coast before it was loaded. To the west, the Strait of Malacca was the only sea passage between the South China Sea and the Indian Ocean, and the kingdom of Srivijaya, based on Sumatra next to Belitung, dominated the straits. A ship from the Gulf passing this way was entering one of the busiest and most closely controlled waterways of the ninth-century world.
How is the wreck dated?
The date comes from the cargo rather than the hull. One Changsha bowl is inscribed with the 16th day of the seventh month of the second year of the Baoli reign, which corresponds to 826 AD. Radiocarbon dating of star anise found in the wreck is consistent with that date, so the sinking is usually placed around 830 AD. Some objects aboard were older; an eighth-century bronze mirror with a cosmological design was probably an heirloom or a piece of old stock, a reminder that the newest object, not the oldest, is the best guide to when a ship went down.
The cargo as a map of ninth-century trade
The detailed catalogue is on our Belitung treasure page, but the make-up of the cargo is evidence in its own right. The bulk consisted of Changsha bowls from the Tongguan kilns in Hunan, packed in straw cylinders or large storage jars, alongside white wares and early blue-and-white dishes from northern China and Yue celadon from Zhejiang. The bowls were decorated for buyers far away: Buddhist lotus symbols, Central Asian and Persian motifs, Koranic inscriptions, and green-splashed wares of a kind popular in Iran. That points to a well-organised export industry producing goods for markets from Southeast Asia to the Gulf.
The luxury pieces tell the same story at the top of the market. The gold cup, the largest Tang gold cup known, shows musicians and a dancer, and its thumb plate carries two curly-haired men who do not look Chinese. Spices, resin and metal ingots carried as ballast filled out the hold. John Guy of the Metropolitan Museum of Art called the find “the richest and largest consignment of early ninth-century southern Chinese gold and ceramics ever discovered in a single hoard.”
Why is the excavation still controversial?
Critics object to three things: the speed of the work, the commercial funding, and the sale of the finds. When the Smithsonian’s Arthur M. Sackler Gallery prepared to host the American debut of the collection in 2012, a group of archaeologists and anthropologists linked to the National Academy of Sciences, including the former Smithsonian head Robert McCormick Adams, wrote to the Smithsonian’s secretary, G. Wayne Clough, on 5 April 2011, warning that the show would “severely damage the stature and reputation” of the institution. On 28 June 2011 the museum postponed it. James Delgado, then director of maritime heritage at NOAA, argued that selling any finds from an excavation “betrays the most basic aspects of research,” though he suggested the Smithsonian could use the show to teach the public about the problem rather than cancel it.
Defenders reply that the excavation was legal under Indonesian law, carried out at the government’s request, and done two years before the 2001 UNESCO Convention on the Protection of the Underwater Cultural Heritage set international standards. Lu Caixia of the Institute of Southeast Asian Studies in Singapore wrote that the hull was properly recorded, the cargo kept together and the finds conserved, catalogued and published, and that it was hard to see how the project could have been paid for without commercial involvement. Victor Mair of the University of Pennsylvania argued that cancelling the show would deny the public and researchers access to the collection. Clough later noted that the United States had never signed the UNESCO convention and said it was time for the profession itself to decide what to do about such cases.
The historian Natali Pearson’s 2023 book Belitung: The Afterlives of a Shipwreck reframes the argument. Using first-hand accounts from the salvagers, she treats the dispute itself as part of the wreck’s history: who owns “shared heritage,” why several major museums declined to host the collection, and how the objects have been used to tell stories about Singapore and the maritime Silk Road. She also records that some Western archaeologists have gone as far as arguing that the objects should be returned to the sea, a measure of how deep the disagreement runs. Her conclusion is that the wreck’s value lies in the relationships between site, objects and people, not in the romance of treasure.
Where did the finds go?
After desalination and conservation in a private facility, a process that took six years, the cargo was sold as a single collection in 2005 for about US$32 million to the Sentosa Leisure Group, now the Sentosa Development Corporation, with the government of Singapore. Its first public showing ran from 19 February to 31 July 2011 at the ArtScience Museum at Marina Bay Sands. In April 2015 Singapore announced that the Asian Civilisations Museum would house it permanently, and it is now displayed there as the Tang Shipwreck collection in the Khoo Teck Puat Gallery. Selected pieces travelled to New York in 2017 for an exhibition at Asia Society, from 7 March to 4 June.
The Jewel of Muscat: sailing a replica
The most direct test of the excavation’s findings was to build the ship again. The governments of Oman and Singapore built a replica, the Jewel of Muscat, working with Flecker. Construction began in October 2008 at Qantab near Muscat. The planks and frames were cut from Afzelia africana trees felled in Ghana, steamed one at a time until they would bend, and then sewn into place with coir rope; workers had about two minutes to clamp each hot plank before it stiffened.
The Jewel left Oman on 16 February 2010 under captain Saleh al Jabri, using a kamal, a simple wooden tool for measuring the height of stars, to check latitude. The voyage did not go smoothly, which made it more useful. Becalmed, the ship was towed by the Indian Coast Guard to Cochin, where her fouled hull was scraped and coated with chunam, a mixture of lime and goat fat. On the way to Galle in Sri Lanka strong gusts cracked a mast, which was replaced with fresh teak, and in Galle the canvas sails were swapped for sails woven from palm leaves in Zanzibar to test how they handled. She reached Singapore on 3 July 2010, about 5,000 km from her starting point. From October 2011 the replica was displayed at the Maritime Experiential Museum on Sentosa; that museum closed in March 2020 and its site was redeveloped as part of the Singapore Oceanarium.
The rig was the least certain part of the reconstruction. The wreck preserved no mast or sails, and almost nothing is known of how such ships were rigged beyond pictures suggesting square sails, so the builders made an educated guess: a hand-made canvas mainsail of 81 m² weighing more than 150 kg, and a smaller mizzen on a second mast. What the voyage did show is that a hull held together only by stitching could survive weeks at sea under sail, but that it needed hauling out, cleaning and a fresh protective coating along the way. In that sense the replica tested the excavation’s records as much as the ship: it could only be built because the hull had been measured carefully under water.
How does Belitung compare with other wrecks?
The Belitung dhow went down about 700 years before the Portuguese Bom Jesus and some 900 years after the Roman-era Antikythera wreck, whose own excavation history runs from sponge divers to modern rebreather teams. For other Asian trade wrecks, see treasure found near Minjiang.
What is the Belitung shipwreck?
An Arab dhow that sank around 830 AD off Belitung, Indonesia, carrying about 60,000 Chinese objects.
When was the Belitung shipwreck found?
In 1998, by local fishermen.
How deep is the Belitung wreck?
About 17 m (56 ft), roughly 1.6 km off Belitung Island.
What was the Belitung ship carrying?
Mostly Changsha bowls, plus celadons, early blue-and-white ware, gold and silver.
How was the Belitung dhow built?
Its planks were sewn together with coconut-fiber cord instead of nails.
Where is the Belitung treasure now?
At the Asian Civilisations Museum in Singapore, as the Tang Shipwreck collection.
Why is the Belitung wreck controversial?
It was excavated quickly by a commercial company and the cargo was sold, which many archaeologists criticize.
How old is the Belitung shipwreck?
About 1,200 years old; one bowl aboard is dated 826 AD.
- Wikipedia: Belitung shipwreck; Jewel of Muscat.
- M. Flecker (2001). A ninth-century AD Arab or Indian shipwreck in Indonesia. World Archaeology 32(3), Shipwrecks issue.
- R. Krahl, J. Guy, J. K. Wilson and J. Raby (eds.) (2010). Shipwrecked: Tang Treasures and Monsoon Winds. Smithsonian Institution, National Heritage Board and Singapore Tourism Board.
- N. Pearson (2023). Belitung: The Afterlives of a Shipwreck. NUS Press.
- D. Gibbins: The Belitung wreck (9th century AD).
- Asian Civilisations Museum, Singapore: Tang Shipwreck collection, Khoo Teck Puat Gallery.
Blane Perun has been exploring Earth’s oceans and marine life for over 25 years. Founder of TheSea.Org in 1999, underwater photographer, coral aquaculture pioneer, and explorer of 80+ countries.
About Blane Perun →