New research suggests that the Madagascan site of Teniky may have been built by a community of Zoroastrian and Iranian origins.
Figure 1. Rock-cut Chamber at Teniky, a) a panoramic photograph from the entrance to the chamber, b) photographs from within the chamber showing the rock-cut chambers along the walls and the central pillar. Credit: Schreurs et al. 2024
During the first half of the twentieth century, visitors to an inland Madagascan site described coming across vast man-made terraces placed within an Amphitheatre-shaped valley (cirque). In the surrounding steep sandstone cliffs, they came upon rock-cut niches, sandstone walls, and pillared chambers.
This place was called Teniky. Located in the Isalo massif in southern Madagascar, the enigmatic rock structures are a mystery. Who built them, when, and why is not known.
However, archaeologists Prof. Dr. Gido Schreurs and his colleagues investigated the Teniky ruins. Their research led them to discover new archaeological ruins, including further terraces, stone basins, rock-cut niches, and sandstone walls and quarries. Furthermore, their research revealed the structures were built around the 10th-12th century and tentatively suggests the settlers who built it may have been of Zoroastrian origins.
Madagascar and Teniky
Figure 2. Images of some of the rectangular niches carved into a rock shelter and remains of a stone wall. Credit: Tristan Allegro
Madagascar was one of the last big islands to be settled by humans. There is still much debate about the origins of these early settlers and the subsequent Malagasy population. While its inhabitants all have some African ancestry, they also share aspects of their language, culture, and, more recently, genetics with Austronesian, Indian, Arab, and Persian populations.
Archaeologically, the first settlement of the island dates to the first millennium (around 1000 AD). Some arguments have been made for earlier settlement, including the purported presence of stone tools in northern Madagascar dated to 2000 BC or cut-marked bones between 10,000 and 1000 years old. Although these claims are highly disputed, it is broadly accepted that the first settlers in Madagascar arrived between the mid-and late first millennium.
Teniky site is located in Isalo National Park. To reach it, one has to hike 20km across steep canyons and rugged terrain. The site has been described for over 100 years; however, until now, none of the other surrounding structures outside of the fluvial cirque have been reported.
The most well-known and prominent structure is the Grande Grotte, a large rock shelter delimited by sandstone walls. In the 1940s, French naturalists Alfred and Guillaume Grandidier believed the structures were likely made by shipwrecked Portuguese sailors who passed through the area in an attempt to reach a port where Portuguese ships may lay. During their traversal, they stopped at Teniky, where they built the Grand Grotte and the surrounding terraces.
Considering the mammoth task this would have been, especially for some sailors just passing through; this interpretation was later questioned. In the 1960s, a trench was dug at the Grande Grotte, which revealed a few pottery sherds that were determined to be part of a Chinese jar from the 16th century.
This was among the last investigations undertaken at Teniky until 2019. High-resolution satellite images of the site revealed Teniky’s actual extent and motivated Prof. Dr. Schreurs and his colleagues to undertake a detailed archaeological excavation. Ground surveys were conducted to locate, note and describe the various structures identified in the satellite images and determine if any others were located at the site. The aim was to better understand the existing structures and perhaps highlight the origins and culture of those who had built and inhabited the area.
The survey revealed multiple rock-cut niches, previously unreported rock-cut chambers, rectangular and circular structures, stone basins and more stonewalling. Also discovered were six stone tombs; although they had all previously been looted and subsequently destroyed, they likely belonged to the Sakalava ethnic group. This group had lived in the area until the Bara drove them out around the 19th century, and they still live in the Isalo National Park region today. Thus, the stone tombs, while interesting, are unlikely to have anything to do with the Teniky site.
Results of the Survey and excavation
Figure.3. Panoramic photograph of rock-cut niches at Teniky. Credit: Schreurs et al. 2024
The survey discovered new features and investigated some of the older, already-known structures, including the Grande Grotte and its sister, the Petit Grotte. The sandstone walls located in the Grande Grotte and photographed by Alfred and Guillaume Grandidier in the 1940s were located. However, they have since partially collapsed. Along the blocks used to construct the wall, archaeologists noted names and dates (graffiti), some modern and others dating back to the late 19th century. South of the Grande Grotte, various circular openings were carved into the cliff, which was around one meter above the ground. Circular recesses around them suggested they had once been closed off with a wooden or stone slab.
Figure 4. Circular niches with recesses around the openings that may have been closed off with wooden or rock slabs. Credit: Schreurs et al. 2024
At the Petit Grotte, the rock-cut chamber was described. It was held up by several massive stone pillars, and the sides contained carved benches. It, too, had niches carved into the surrounding cliffs. These niches were 50–60 cm wide and usually situated about 1.2 m above the ground.
Other structures found across Teniky include over 35 circular niches and around 10 rectangular ones, all of which were always carved at least a few dozen centimetres above the ground. Vertical walls with pillar-like shapes were carved into the rock face, and boulders that looked a lot like small model versions of the great rock-cut chambers were cut. One even displayed a gable-like roof.
Figure 5. A heavily eroded rock-cut boulder with a slightly gabled roof. Credit: Schreurs et al. 2024
Further discoveries included stone basins with hollows potentially used to hold water, fire, or some other substance. More stone walls, although with an accompanying foundation, were not noted. The stones used in the construction of the wall were likely sourced from a nearby quarry, also identified during field surveying.
The excavations revealed very little. While some charcoal flakes were found, it was impossible to determine if these resulted from man-made fire or natural fire. Furthermore, the few recovered animal bones were unidentifiable and showed no traces of man-made cut marks. Thus, it was not possible to determine if these were the result of human accumulation or natural deposits, the result of animals dying at or near the chambers.
Finally, ceramic sherds were retrieved. These, along with the charcoal, gave an indication of when Teniky was built.
Four radiocarbon dates were obtained from charcoal recovered during excavation. They suggest that the sandstone walls were first erected around the 10th to mid-12th centuries. The pottery sherds recovered during excavation supported these findings.
Figure 6. Chinese and Southeast Asian ceramics from a) eleventh/thirteenth-century, b-c) twelfth-century, and d-i) thirteenth/fourteenth-century.
The sherds were found to be of Chinese and Southeast Asian origin. According to Bing Zhao the Southeast Asian ceramic sherds were manufactured around the 11th and 13th centuries, while the Chinese celadon and Southeast Asian stoneware were made around the 11th and 14th centuries.
Interpreting the Finds
Figure 7. Stone Basin with some rock-cut niches in the background. Credit: Schreurs et al. 2024
Finding individual pieces of evidence is great, but now the archaeologists have to interpret them. Considering the dimensions of the chambers and niches and the effort that went into carving them, they were determined unsuitable for living and too much effort for a simple storage space. Therefore, the researchers hypothesised a ritual function.
The lack of material culture further supported this. One would assume that if people had lived here or used it to store things, one would find the remnants of daily life and stored objects.
While the chambers may have been ritual, the terraces, which span over 30 hectares were more likely associated with settlement. The presence of large carved boulders and stone basins at level with the terraces suggests that they were built contemporaneously with the rock-cut structures.
This indicates that the stone structures, terraces, and chambers were all built around the same time, likely by a specific group of people who arrived in the area around 1000 years ago. However, the limited excavations so far make it difficult to say how long they stayed at Teniky or when and why they abandoned the site.
Perhaps, something could be said about who made the structures.
Teniky’s rock-cut architecture is not known anywhere else in Madagascar or even the adjacent East African coast. What can be ruled out is that the structures were made by shipwrecked Portuguese sailors. Portuguese ships only began sailing in the Indian Ocean around 1498, which postdates the proposed construction of Teniky by a few centuries. However, while the Portuguese did not build the site, it is possible that they came across it while traversing the island.
It is possible that no similar architecture like that at Teniky exists elsewhere because the people who settled here developed their own unique culture and architecture after arriving on the island. However, considering the relatively short period of time between the arrival of the first Madagascan settlers and the proposed construction of Teniky, it is unlikely. This is because incoming settlers usually take with them and continue practising the traditions and cultural customs of their place of origin. These may evolve with time but are unlikely to be completely overhauled in favour of something radically new upon arrival in a new space.
Thus, if similar structures could not be found in Madagascar or East Asia, the researchers had to look elsewhere. Considering the contribution of Asia, Indian, Austronesian, Arab, and Persian ancestry, culture, and language to the Madagascan population, perhaps Teniky’s origins could be traced back to these places and people as well.
Iran and the Zoroastrians
Figure 8. Pictures of rock-cut niches in the hills north of Sirāf, Iran (a-b) and rock-cut niches in Teniky, Madagascar (c-d). Credit: Schreurs et al. 2024. Images of Siraf niches courtesy of Ali Aghajanzadeh.
The Fars region in southern Iran was once host to the port city of Sirāf. Sirāf thived in the 9th and 10th centuries; ships that entered its entrepots brought merchandise from the Middle East to India, China, the Red Sea, East Africa and South-East Asia. However, it began to decline in the late 10th century due to a number of events from which it could not fully recover, including a series of earthquakes in 977, the rise of a rival port in 1010 (Qais) and the collapse of the Biiyid Dynasty around 1055. Archaeologist David Whitehouse, together with the British Institute of Persian Studies, carried out a series of excavations between 1966 and 1973 at Sirāf. While there, he noted hundreds of niches of different sizes cut into the steep cliff walls around the city. He speculated these to be ossuaries (type of tomb) used by the Zoroastrians.
Zoroastrianism is one of the oldest monotheistic religions in the world. It was a dominant state religion in the Persia Sasanian Empire between 224–656 AD. After the conquest of the Sasanian Empire by the Arabs during the mid-7th century Islam was imposed. For a while, it coexisted with Zoroastrianism; however, gradually, more conversions to Islam meant only a few Zoroastrians still lived in Sirāf by the 10th century. Some Zoroastrians left Iran and settled in India during the 8th century, where they were known as Parsis (i.e. people of Pars or Fars). Today, a few small groups of Zoroastrians still exist in Iran.
Prof. Dr Schreurs and his colleagues noted various similarities between the Sirāf Zoroastrian site and Teniky, including rock-cut benches within chambers that served as niches, the niches being closely clustered together, giving them the appearance of pillared chambers, openings of niches being rectangular or circular, always being situated above the ground, and recesses cut around some niches allowing them to be closed off. The main differences are that the rock-cut niches in southern Iran are never associated with stone walls, nor do they have vertical slabs placed on the ground in front of them. However, this may be accounted for by the fact that limited archaeological work has been done on the niches at Sirāf and in southern Iran. Thus, these features may well exist but have simply not been documented or discovered.
The Zoroastrians did not bury the dead in the ground as the body was seen as polluting. Instead, they were left in places of display above ground where they could naturally decay or be consumed by dogs and vultures. These areas of display or niches are called dakhmas in Pahlavi. The bodies of the deceased were laid in these large rectangular niches until they had become nothing more than bones. After decomposing, the remaining bones were transferred to smaller circular niches which could be closed off, called astōdans. This would protect them from rain and prevent the deceased’s remains from polluting the earth.
Figure 9. Image of the some Siraf rock-cut niches. Credit: Ali Aghajanzadeh in Schreurs et al. 2024
Prof. Dr. Schreurs and his colleagues tentatively interpret the Petit Grotte and Grande Grotte as chamber tombs, with their carved stone benches serving as ossuaries. Meanwhile, the carved boulders around the slope of Teniky may have helped dry or store bones. The niches with recesses in them may have been astōdans used to store the bones once they were completely dry. Differences in size and shape may have indicated the social status of the deceased.
But if the niches once held human remains, why are they all empty now?
One would expect bones, especially denser bones and teeth, to be found in at least some of the niches if they were, in fact, used to store bones in the past. While one could speculate that animals and the elements may have destroyed the bones after the site was abandoned, what about those stone niches that could be sealed off with the express purpose that no animals, rain, or weather would accidentally disturb them?
Steve Goodman, an American field biologist at the Field Museum of Natural History in Chicago who has conducted research in Madagascar for over 30 years may have an explanation for the absence of bones, which he relayed to Prof. Dr. Schreurs,
‘It is certainly possible that the human remains were removed by subsequent people of the area, perhaps with the focus of the material being used for different types of “black magic”, which still goes on today at some of the Bara tomb sites’.
The similarity in architecture thus may point to a Zoroastrian origin of the Teniky structures; however, the stone basins may provide further links. The stone basins at Teniky are similar to the tables, platforms, and stone basins used by Zoroastrians during ritual ceremonies. They held fire and water, both of which were important for ritual purity. Considering some of the stone basins were found in front of the stone niches or carved boulders at Teniky, they were likely connected to whatever was stored in these places.
But is there a connection between Madagascar and Persia, one the Zoroastrians may have utilised to get to Madagascar?
Sadly, only a few primary sources discuss the East African coast between the first and second millennium AD, and of those, fewer mention the island of Madagascar.
However, there are some accounts that indicate Persian sailors had contact with Madagascar. Including the accounts by 10th-century Persian sailor Buzurg Ibn Shahriyār, who makes mention of contacts between Perisan and Madagascar. Furthermore, archaeologically speaking, southern Iranian ceramics dating to the 9th and 12th centuries have been found at various sites in Madagascar, providing both literary and physical proof of a Persian-Madagascan contact.
While archaeologists can only speculate at the moment, it is possible that at least one group of Zoroastrians, whether due to the collapse of Sarif or pressure from the ever-more-influential Islam, chose to leave their home in southern Iran. However, unlike Parsis, who ended up in India, the builders of Teniky found themselves in Madagascar.
Here, they built their necropolises and continued to engage in the Indian Ocean trade, at least to some degree, as evidenced by the Southeast Asian and Chinese ceramics, until, for whatever reason, they abandoned the site.
…
Around 1000 years ago, an unknown group of people settled in the Isalo massif in southern Madagascar. There, they carved large stone chambers and rock niches, built stone terraces and erected stone walls unlike anything else found in Madagascar or the Eastern African coast. However, they were no isolated community cut off from the rest of the world, as they procured ceramics from Southeast Asia and China. Recent excavations at the site of Teniky, where this group of people lived, not only discovered how much larger the site is but also proposed that the original inhabitants may have had roots in Iran. They tentatively suggest that a Zoroastrian community may have lived in Madagascar around the turn of the first millennium AD and constructed a necropolis similar to those found in their homeland.
However, many questions still remain unanswered, such as when exactly they arrived, why they came to Madagascar, what caused them to abandon the site, and how they lived and interacted with other populations across the island.
Madagascar is around a 4-hour flight from where I live; if I ever get the chance, I would love to visit the Isalo National Park. If that ever happens, I wonder what new revelations will have been made about the site. Will further excavations prove or refute this initial hypothesis? I wonder. What do you think? Do you know of any architecture that is eerily similar to that at Teniky? Or do you have some other theories that could suggest the structure’s original use?
References
Schreurs, G., Allegro, T., Rouvinez, M., Radimilahy, C., Raharinoro, J., Fanny Sabe, N., Rakotoarisoa, J.A., Rakotomavo, L., Rakotondrazafy, N.A., Randrianarivelo, N. and Szidat, S. (2024) ‘Teniky: enigmatic architecture at an archaeological site in southern Madagascar’, Azania: Archaeological Research in Africa, pp. 1–44. doi: 10.1080/0067270X.2024.2380619.
Whitehouse, D. (1970) ‘Sirāf: A medieval port on the Persian Gulf’, World Archaeology, 2(2), pp. 141–158. doi: 10.1080/00438243.1970.9979471.
Story by Sandee Oster
Comments