The Belitung shipwreck (also called the Tang shipwreck or Batu Hitam shipwreck) is the wreck of an Arabian dhow that sank around 830 AD. The ship completed its outward journey from Arabia to China but sank on the return voyage from China, approximately off the coast of Belitung Island, Indonesia. The reason the ship was south of the typical trade route when it sank remains unknown. Belitung lies southeast of the Singapore Strait, approximately away, a secondary route that was more common for ships traveling between China and the Java Sea, which is south of Belitung Island.
The wreck has provided archaeologists with two major discoveries: the largest single collection of Tang dynasty artefacts found outside China, known as the "Tang Treasure", and the Arabian dhow itself, which offers new insights into the trade routes between China and the Middle East during that period. The treasure has been preserved as one collection, and efforts during excavation to maintain the integrity of the site and its cargo have produced detailed archaeological evidence. This evidence has provided new knowledge of the shipbuilding techniques of the time, as well as insights into the nature and style of the traded artefacts, shedding light on the trade between these two regions.
Currently, the Tang dynasty treasures recovered from the Belitung shipwreck are on permanent display at the Asian Civilisations Museum in Singapore under the name "Tang Shipwreck".
Discovery and route
Discovery The wreck was discovered by local fishermen in 1998 in the Gelasa Strait, at a depth of .The Belitung Shipwreck - SEAArch - Southeast Asian Archaeology – Archaeology and Cultural Heritage of Southeast Asia The rights to the site were purchased from local fishermen, and a license for excavation was awarded to a local Indonesian company. Tilman Walterfang and his team at Seabed Explorations subsequently financed and conducted the excavation under a cooperative agreement with the original salvage company, at the request of the Indonesian government, which provided security for the site through the Indonesian Navy. The excavation took place over two expeditions, beginning in August 1998 and continuing with a second phase in 1999. Seabed Explorations supplied vessels and funded naval operations to secure the wreck site before and during the monsoon season.
Route It is unknown why the ship was so far from its expected route (shown in red on the map to the right), as most ships departing China for Arabia would have traveled through the South China Sea. Typically, they would then turn northwest after passing southern Vietnam, continuing through the Singapore Strait into the Straits of Malacca between Peninsular Malaysia and Sumatra. Belitung lies some distance from this usual route, making the ship's presence in this area puzzling. The island is southeast of the Singapore Strait by approximately , and this secondary route is generally more common for ships traveling from the Java Sea, located south of Belitung Island, to the Strait of Malacca, which lies around north of the island.
Ship and construction
The shipwrecked dhow was approximately wide and long and is notable for two reasons: it is the first ancient Arabian ship to be found and excavated, and its planks were sewn together using a thin rope made of coconut fibers, rather than the traditional pegs or nails used in Arabia in later centuries.
The wreck timbers were buried under sediment that preserved the wooden remains, preventing them from being consumed by marine worms. Shipwrecks of this age are rare finds, and this particular one was in such good condition that much of the hull was preserved. This has provided valuable insights into the construction techniques of ships from this period—an unprecedented discovery, as no other Arabian ship of this type has been found with its cargo intact.
Sections of the original timbers were well-preserved enough for scientists to analyze them and identify some of the wood types used. It is possible that the ship was built in western Asia and later purchased by Arabian merchants for use on the Oman-to-China route; the cargo includes numerous artefacts with Arabian influences.
Construction techniques
The ship was constructed around a long keel with a thickness of , which is believed to have survived intact. The front of the ship featured a 61° angle of rake at the bow, where the stem post was joined to the keel using mortise and tenon joints, secured by a diameter rope. The hull planks were stitched to the frames and keel through holes spaced apart. Additionally, the boat included a keelson for added strength, resting on the half-frames.
Excavation archaeologist Michael Flecker compared the wreck's construction to contemporaneous vessel types and found the closest parallels in the lashed-lug boats of the Austronesian peoples in Maritime Southeast Asia, such as the Pontian boat from Pahang, Malaysia (c. 260–430 CE), and the balangay boat burials from Butuan, Philippines (c. 320–1250 CE). However, Flecker noted that even these similarities were limited, as lashed-lug construction typically involved edge-joined planks with dowels and lashed protruding lugs, whereas the Belitung wreck featured fully through-stitched planks without dowels. He ultimately concluded that the wreck was an Arabian (or possibly Indian) vessel based on hull form, stitching technique, timber species, and other features. Flecker noted that fully stitched boats were known from regions as far apart as the African coast, Oman, the Red Sea, the Indian coast, and the Maldives.
Roman sources, such as the 6th-century historian Procopius, also reference similar boats with planks stitched together, used in the "Indian Seas". Although no Arabian ships of this type had been previously discovered, references to them exist in historical texts, including the late-Tang ("Strange Things Noted in the South"). According to John Guy, a curator at the Metropolitan Museum of Art, this text "describes the ships of foreign merchants as being stitched together with the fiber of coir-palms and having their seams caulked rather than using iron nails to secure their planks." This method is distinct from the metal fastenings seen in later shipbuilding practices.
Wood types Samples of wood from the shipwreck were sent for analysis at the Forest and Forest Products division of the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Australia. The analysis was conducted by Jugo Ilic, a wood identification specialist at CSIRO. Many of the samples were too deteriorated for positive identification, as the lack of remaining cellulose in the wood cells prevented successful analysis.
Several types of wood were identified: teak (Tectona grandis) was used for the through-beams, as it is resilient to the teredo worm (family Teredinidae). The ceiling was made from a Cupressus species, possibly Cupressus torulosa; the stem post was constructed from rosewood of the family Leguminosae (now called Fabaceae) and either the genus Dalbergia or Pterocarpus. Additionally, a wooden box found in the stern area was made from the genus Artocarpus of the mulberry family, Moraceae.
The species used for the hull planks was not positively identified but is thought to be Amoora from the family Meliaceae. The timber for the frames was similarly undetermined, though one frame was probably from Amoora or from the genus Afzelia of the family Fabaceae. Afzelia is notable because its three main species—A. africana, A. bipindensis, and A. pachyloba—are primarily found in a small part of Africa, stretching from the mid-western coast in a narrow band toward the east coast but stopping a few hundred miles short of it.
Based on construction techniques and materials, it was initially speculated that the ship could be of either Arabian or Indian origin, as there was little distinction between ships from these regions during that period. However, the frame's use of a wood species found only in a limited part of Africa adds complexity to this identification. After analyzing the hull form, timber species, and construction methods, Ilic concluded that the wreck was likely of Indian or Arabian origin, with India being the more probable site of construction, though Arabian origins were not ruled out due to the importation of timber for shipbuilding in the Middle East. Flecker later concluded that the wreck was an Arabian ship, asserting in his 2010 chapter in the Sackler exhibition catalogue that "from an analysis of construction methods and materials and hull form, the author has determined that the Belitung wreck is an Arab vessel."
Legacy Current knowledge of the original materials and methods used in constructing this particular Arab dhow largely originates from the shipwreck itself. The Jewel of Muscat reconstruction—a replica built as an exact copy of the wreck—demonstrated that the ship resembles a baitl qarib, a type of vessel still found in Oman today. Within the hull of the shipwreck, large lumps of concretion contained artefacts from the ship's cargo, dating back to the Tang dynasty of China, around 800 AD. This connection has led to the wreck being referred to as the "Tang shipwreck" or "Tang treasure ship".
The ship's timbers and artefacts were publicly displayed for the first time in 2011. The world debut exhibition, showcasing both the artefacts and timbers from the ship, took place at the ArtScience Museum, adjacent to Singapore's Marina Bay Sands. The historical significance of the shipwreck's discovery led to the decision to construct the Jewel of Muscat as a faithful reconstruction of the original dhow.
Cargo
The wreck contained three main types of Chinese wares in the form of bowls: Changsha ware (produced in kilns in Tongguan), which made up the majority of the 60,000 items and was originally packed in either straw cylinders or "Dusun" storage jars; white ware, manufactured in the Ding kilns, which included the earliest known intact underglaze blue and white dishes; and Yue ware from Zhejiang Province. One Changsha bowl bore an inscription with a date: "16th day of the seventh month of the second year of the Baoli reign," or 826 AD, a date later confirmed by radiocarbon dating of star anise found among the wreckage. The cargo showed a variety of influences and markets, featuring designs such as Buddhist lotus symbols, motifs from Central Asia and Persia, Koranic inscriptions, and green-splashed bowls popular in Iran.
The cargo also included a variety of items, such as spice jars (martaban), ewers, inkwells, funeral urns, and gilt-silver boxes. According to John Guy of the Metropolitan Museum of Art in New York, the cargo represented "the richest and largest consignment of early ninth-century southern Chinese gold and ceramics ever discovered in a single hoard." The cargo also included spices, resin, and metal ingots used as ballast. Notable items found include a large gold cup—the largest Tang dynasty gold cup ever found—and a silver flask decorated with a pair of ducks. The gold cup features images of people engaged in various activities, such as musicians and a Persian dancer. The thumb plate on its handle displays two men with non-Chinese features, depicted with curly hair.
Current disposition Tilman Walterfang's company formed a cooperation contract with the original Indonesian salvage company to ensure the cargo was preserved as a single collection rather than sold piecemeal to collectors. Although some looting occurred at the site, especially between excavation periods, Walterfang maintained the integrity of the cargo, allowing it to be studied in its original context. According to Julian Raby, director of the Arthur M. Sackler Gallery, this decision has provided "unparalleled insight into China's industrial capacity and global trade."