Tombs, Landscape and Society in Southern Madagascar - Part IV

Mike Parker Pearson

The 16th-17th century royal centre at Ampotake

Further inland from Talaky on the west bank of the Manambovo is the area traditionally known as the earliest settlement, after Anjampanorora and Montefeno, of the Andriamañare, the royal clan of the Tandroy. This is almost certainly the 'town' of 'Murnanzak' (Andriamananjaka), the nephew of Deaan Crindo (Andriankirindo) and the rightful heir to the Tandroy throne as recorded in the early 18th century by Robert Drury (1729). Today most of this ancient settlement is underneath a large sacred forest containing Andriamañare tombs. However, prospection along the northern edge of the forest revealed the presence of graphite wares probably dating to the 17th century if not earlier. No settlements dating to before the 17th century were found on either side of the Manambovo estuary and there was only a handful of mainly small 17th century settlements in the area.

In contrast, a large number of settlements with 18th century pottery were found on both banks, including one which had been covered by sand dunes and is now partially revealed in an unusually good state of preservation. During the 19th century there was a further expansion in the number of communities, especially on the previously largely empty east bank around the limestone massif of Faralambo. It was here that the ancestors of the newly arrived Afomarolahy clan were settled by the Andriamañare king at the end of the 18th century.

Ancient capitals of the Mahafaly on the Menarandra River

A first visit was made to locate other royal centres of the Mahafaly kingdom on the Menarandra river, to the west of Androy along the river's west bank. Although this region is largely inhabited by communities of Karimbola, it was formerly one of the five kingdoms of the Mahafaly. Robert Drury recounts his involvement as a royal slave in a war between the Tandroy and the Mahafaly when an army of Tandroy and people of the Fiherenana (the area north of Tulear on the west coast) invaded Mahafaly and defeated them in a battle on the west bank of the Menarandra near the royal capital. In 1980 and 1983 Chantal Radimilahy fieldwalked the ancient Mahafaly capital of Ampasimahanoro, occupied between the mid-18th and mid-19th centuries, recovering pottery, gunflints and stone pipes. North of this we found two other large settlements of late 17th-early 18th century date. The earliest was at Ankilisoa, immediately south of the area known as Andriamanda. The later one was located further north, just north of the river's junction with a tributary, the Sakatovo, and east of a lake known as Fenoarivo. On the south side of the Menarandra-Sakatovo fork, perched on cliffs over 30m high, is another large settlement dating from the 13th-15th centuries. Its surface finds include two pieces of Chinese celadon. The absence of any 15th-17th settlement may be explained by Flacourt's observation that the Mahafaly were nomadic pastoralists in this period.

Elephant Bird (Aepyornis)

Early Occupation and the Elephant Bird

The Elephant Bird, identifiable as the roc or rukh in the stories of Marco Polo and Medieval Arab voyagers, was the largest bird that ever lived. This enormous rattite was flightless and appears to have died out around the same time as the other megafauna of Madagascar, such as the giant lemurs and pygmy hippopotami. The dunes of the south coast are often carpeted with its broken eggshells and very occasionally complete unbroken eggs are found. Curiously, the skeletal remains of these birds are extremely rare and little is known of the taphonomic processes leading to their disappearance.

Whilst eggshells are found in profusion along the coast, bones are not. This may, in part, be due to the birds having been seasonally migratory, moving to the south coast during the nesting season and then spending most of their time to the north or inland. In 1962 Professor Pierre Vérin excavated two shell middens at Talaky, in the dunes on the west side of the Manambovo rivermouth, where he found Aepyornis eggshells in association with pottery and human occupation radiocarbon dated to the 10th-12th centuries AD. We resurveyed this area in 1995 and identified more extensive and possibly earlier shell middens here. This season we hoped to excavate some of these sites to clarify whether the early inhabitants were eating the eggs as surmised by Vérin. The likely earlier midden turned out on excavation to be an ancient sand dune without any evidence of in situ human occupation. We also excavated nine trenches through two middens dating, on the basis of pottery style, from about the 10th to the 15th centuries, and one trench through a 19th century midden. Aepyornis shell was found in probable association with the earlier middens in three instances and certainly in situ in two cases. A geomorphological survey of the surrounding dunes also located two relatively recent complete but broken eggs. A major programme of radiocarbon dating is now planned. The eggshells and charcoal from the middens should give us the first reliable insight into the interrelationship of the giant bird with early occupation in southern Madagascar and will also provide the first detailed radiocarbon chronology for the four main types of the earliest ceramic styles. Radiocarbon dating of eggs from the dunes, coupled with OSL (optically stimulated luminescence) dating of associated sediments will also provide dates for the stratigraphically latest eggs found outside the middens.

The 1997 field season

A five-week field season was completed in August and September 1997. The main aims were to investigate the relationship between the extinct Elephant Bird (Aepyornis) and the earliest human settlements in the south of Madagascar, to investigate the location and context of 16th-17th century Tandroy royal centres and earlier sites around the rivermouth of the Manambovo, to investigate the iron cannons noted by Defoort in 1913 and associated with a shipwreck south of Erada, and to discover the ancient centres of the Mahafaly along the west bank of the Menarandra river. Despite initial mechanical difficulties with the land rover, this season's research was extremely successful.

Examination of one of the cannons

The shipwreck of the Degrave?

Around the turn of the century the ethnographer Emile Defoort noted the survival of two iron cannons on the beach about 20 km southwest of Ambovombe. He was certain that these were the remains of the English East Indiaman, the Degrave, which brought the young Robert Drury to Madagascar in 1703, prior to the Tandroy's massacre of the crew and the enslavement of Drury.

In 1995, part of a bronze bell was recovered by snorkel divers from the actual wreck site, adjacent to one of the cannons but just outside the reef. There are apparently the remains of at least four cannons and an anchor here. The bell fragment has no identifying marks but was made before c.1750 and is probably English. The cannons are 'finbankers', a form produced by the Swedes, Dutch and English, though one cannon is probably English. Both are considered to date broadly to the late 16th-17th centuries and one assessment dates them to 1665-1675. The dating of the cannons and the bell concurs well with our impression, based on Drury's and Benbow's accounts, that this wreck site is that of the Degrave, built in 1698. This discovery has allowed us to reassess the likelihood that a small copper alloy Bodhisattva from the western Himalayas, examined in 1993 at Benonoke, probably originally derived from the Degrave.


Our team were Karen Godden, Georges Heurtebize, Ramilisonina, Victor Razanatovo, Retsihisatse, Jean-Luc Schwenninger and Helen Smith. The fieldwork was funded by the National Geographic Society, the Society of Antiquaries and the British Academy. The satellite remote sensing project was funded by NERC and carried out with Simon Garrod and Chris Clark.


  • Battistini, R., Vérin, P. & Rason, R. 1963. Le site archéologique de Talaky. Annales Malgaches 1: 111-27.
  • Clark, C., Garrod, S. & Parker Pearson, M. Forthcoming. Landscape archaeology, deforestation and remote sensing of southern Madagascar. International Journal of Remote Sensing.
  • Drury, R. 1729 [1890]. Madagascar: or Robert Drury's Journal during fifteen years captivity on that island. London: Fisher Unwin.
  • Flacourt, E. de. 1661. Histoire de la grande isle de Madagascar, composée par le sieur de Flacourt, avec un relation de ce qui s'est passé les années 1655, 1656, et 1657 non encore veue par la première impression. Paris: Clouzier.
  • Heurtebize, G. 1986a. Histoire des Afomarolahy (extrême-sud de Madagascar). Paris: CNRS.
  • Parker Pearson, M. 1996. Re-appraising Robert Drury's Journal as a historical source. History in Africa 23: 1-23.