Under the Old Fort of Zanzibar: exploring the origins of Stone Town

Timothy Power and Mark Horton return to the Old Fort of Stone Town, Zanzibar, to discuss the site’s Swahili origins.

The UNESCO World Heritage Site of Stone Town, Zanzibar, is today renowned for its elaborate 19th-century merchants’ houses. It is often assumed that the development of the town itself was the product of Omani Arab colonialism, when, in the mid-19th century, Sayyid Saïd made Zanzibar the capital of his Indian Ocean empire (see CWA 116). However, our recent archaeological work at the Old Fort located close to the seafront of Stone Town suggests that the origin of this settlement is nearly a thousand years older, dating back to the golden age of Islam and the emergence of the Swahili civilisation.

above Archaeological trenches in the courtyard of the Old Fort in Zanzibar revealed evidence for the origins of
Archaeological trenches in the courtyard of the Old Fort in Zanzibar revealed evidence for the origins of Stone Town.

The origins of Stone Town

We discovered that the original shoreline lies some 2m below the present courtyard of the Old Fort. The beach sand scattered with rounded boulders is evocative of the same tropical beaches that so dazzle international tourists today. Alternating bands of dark sand with ceramic sherds and clean white sand tell of episodes of domestic dumping and shore agglomeration. The source of this refuse was most likely a nearby settlement.

About 20m inland from this now deeply buried beach, we found a large hollow scoop cut into the sand, most likely a broad, open well. These are quite common, even today, around the Indian Ocean, and work on the principle that fresh water floats over the heavier seawater. They often have a short life, as the salty water can get easily mixed – we found this ancient ‘well’ choked by sand. The settlement then expanded over the beach with a series of reddish-brown clay floor surfaces from daub huts, cut by ‘robber’ pits associated with daub recycling, all suggesting quite intensive settlement activity.

While the ceramic assemblage is dominated by locally produced Swahili cooking pots, significant numbers of vessels came to the settlement via the monsoon trade winds, and point to extensive international contact. Handis from India, sgraffiatos from Persia, and stonewares from China, carried over thousands of miles, seemed to be the everyday vessels of the Zanzibar Swahili. These imported ceramics allow us to date the beginning of occupation under the Old Fort – and so the origins of Stone Town itself – to the late 10th century.

ABOVE Sgraffiato ceramics from south-east Persia, made between the 11th and 13th centuries. ABOVE RIGHT Traditional Swahili daub houses at the modern village of Unguja Ukuu.
Sgraffiato ceramics from south-east Persia, made between the 11th and 13th centuries.

This date is important because it corresponds to the abandonment of Unguja Ukuu, the main Indian Ocean port on Zanzibar between the 8th and 10th centuries, ten miles down the coast. It is also the date for the decline of Siraf, now in Iran, and one of the most important ports for the Indian Ocean trade. Siraf suffered a massive earthquake in AD 977 and many of its merchants moved out. Stone Town in faraway Zanzibar may have been associated with a major reconfiguration of trade networks and settlement patterns at this time.

A mosque on the beach

In the following centuries, the village of daub huts transitioned into a town of stone houses.

At one point we were able to trace a sequence of buildings: Building 1 was constructed of coral blocks set in a strong lime mortar, sometime in the 12th to 13th centuries. Since it is generally thought that domestic stone architecture spread through the Swahili Coast near the start of the 14th century, this building is unusually early.

Traditional Swahili daub houses at the modern village of Unguja Ukuu.

While we don’t have the full plan, its orientation east–west/north–south towards Mecca indicates it was a mosque. After a few hundred years, it was carefully dismantled prior to the construction of Building 2, which included a plaster column base set in clean white sand. Column bases of the same type are commonly found in medieval Swahili mosques, suggesting that the earlier mosque was replaced with a more impressive structure.

Such an interpretation is strengthened by the discovery of a carved stone mihrab block (the niche in a qiblah wall facing Mecca) reused in a later Portuguese wall. Similar blocks were used in the construction of the famous decorated mihrab of the Kizimkazi Mosque, in the south of the island, dated by inscription to AD 1107. This piece of architectural spolia probably belonged to an abandoned or destroyed mosque that may once have stood on the beach.

The Black Death

For reasons that are not entirely clear, the putative mosque was abandoned and fell into ruin, and the area was taken over for domestic use. A shell midden was found in the ruins, and this was followed by the construction of Building 3. The walls were thinner and less well made than the previous buildings, which together with the discovery of a cooking pot reused as a bread oven, suggests a typical Swahili stone house.

LEFT The medieval occupational sequence exposed in an archaeological trench. The remains of four buildings were found spanning the 12th/13th to 16th/17th centuries.
The medieval occupational sequence exposed in an archaeological trench. The remains of four buildings were found spanning the 12th/13th to 16th/17th centuries.

Ceramics from the final occupation and subsequent collapse of Building 3 include far fewer imports, implying that Stone Town’s participation in Indian Ocean trade was retreating. Imports such as green monochrome glazed ware from Persia and celadons from China provide a late-14th- to 15th-century date for the third phase. We might tentatively suggest that the decline in settlement and trade was linked to the Black Death, which not only devastated Europe, but also spread across the Indian Ocean and Africa.

Unfortunately, the stratigraphic sequence was at this point interrupted by levelling of the site to make way for the terminus of an early 20th-century narrow gauge railway, known as the Bububu Railway (named because of the sound it made!). Nevertheless, truncated rubbish pits and Muslim grave shafts were found, which can be placed in the final phase of the medieval Swahili town. This apparent downturn in the fortunes of Zanzibar – and indeed many of the Swahili ports – may go some way to explaining the success of the Portuguese in establishing their control of the East African coast in the early 16th century.

Timothy Power is an Associate Professor of Archaeology at the UAE University and Mark Horton is a Professor of Archaeology at the Cultural Heritage Institute of the Royal Agricultural University. Their research specialisations bring together the Arabian Peninsula and East Africa as part of a shared interest in the archaeology of the premodern Indian Ocean world. The 2022 field season was funded by the Dhakira Center for Heritage Studies at New York University Abu Dhabi and carried out in collaboration with the Ministry of Tourism and Heritage, Zanzibar.
IMAGES: Tim Power