Early Berbera: Malao, Po-pa-li and the 'land of Berbera' before 1300

In short

Berbera has the best natural harbour on the Somali shore of the Gulf of Aden. Its early history is known only from foreign writers and from ruins that have not been excavated. This page gathers what each source says about it before 1300. For the name "Berbera" itself, which Arab and Greek writers used for the whole coast, see Greek and Roman writers on the Somali coast.

Malao in the Periplus, 1st century AD

The Greek merchants' guide called the Periplus of the Erythraean Sea describes Malao, about 800 stadia east of Avalites (Schoff, The Periplus of the Erythraean Sea, 1912, §8, p. 25):

"After Avalites there is another market-town, better than this, called Malao… The anchorage is an open roadstead, sheltered by a spit running out from the east. Here the natives are more peaceable. There are imported into this place the things already mentioned, and many tunics, cloaks from Arsinoe, dressed and dyed; drinking-cups, sheets of soft copper in small quantity, iron, and gold and silver coin, not much. There are exported from these places myrrh, a little frankincense (that known as far-side), the harder cinnamon, duaca, Indian copal and macir, which are imported into Arabia; and slaves, but rarely."

Schoff wrote that "Malao is the modern Berbera" (same source, notes p. 79). Ptolemy, about 150 AD, also lists a "Malao market place" between Avalites and Mondu (Stevenson's translation of Ptolemy's Geography, 1932, p. 107). See the Somali ports in the Periplus.

Malao: Berbera or Siyaara? Who says what

A Spanish team that surveyed the Berbera coast in 2018–2020 questions the usual reading. The Periplus calls Malao's anchorage an "open roadstead" sheltered by a spit or headland "running out from the east". The team writes that this "does not fit modern Berbera at all, which has a closed, calm roadstead", and that it fits Siyaara, about 30 km east of Berbera, better: Siyaara has a prominent rocky headland, the only one between the Djibouti border and central Somaliland, and wells with drinking water all year. They suggest looking not for one spot but for a stretch of about 100 km of coast, from the mouth of the Issutugan river to Siyaara, where different landing places may have been used (González-Ruibal, de Torres et al., "Nomads Trading with Empires: Intercultural Trade in Ancient Somaliland in the First to Seventh Centuries CE", American Journal of Archaeology 126.1, 2022, pp. 119–120, open access).

What they found on that coast (same source, pp. 120–126):

• Siyaara. Pieces of three Himyarite (South Arabian) jars, dated loosely from the 1st millennium BC to the mid-1st millennium AD, and several large pre-Islamic cairns on the shore, placed "to be seen from the sea or the coast". Siyaara was later the site of a large seasonal fair, at its height between the 11th and 16th centuries.
• Ceel Gerdi, 22 km east of Berbera. Six shell middens run for 6 km along the shore. Three are Neolithic; three hold imports of the 5th–7th centuries AD, among them a Late Aksumite jug from eastern Tigray, a Himyarite jar and a dark glass bangle of a type common in Egypt and the Levant. The team reads them as short-stay camps of small nomad groups who herded, fished, gathered shellfish and probably hunted. A camel bone from one midden is, they write, the oldest found so far in Somaliland.

They conclude that the coast's trade at this period ran mainly through a regional network linking the Horn with South Arabia, and that the nomads of the Berbera coast were "not the typical specialized pastoralist society" but "mobile communities with a mixed economy" (pp. 126, 132).

Po-pa-li and Pi-p'a-lo in the Chinese record

• Mid-800s. Duan Chengshi's Youyang zazu describes "the land of Pa-pa-li… in the south-western Ocean". Its people "do not eat any cereals, but they eat meat; more frequently even they prick a vein of one of their oxen, mix the blood with milk and eat it uncooked". They wrap "a sheep's skin" round their waists. "The people of this country make their own countrymen prisoners whom they sell to strangers." Its products are ivory and ambergris (Hirth and Rockhill, Chau Ju-kua, 1911, p. 128, note 1).
• 1225. Zhao Rugua writes that "the country of Pi-p'a-lo contains four cities; the other (places) are all villages which are (constantly) at feud and fighting with each other". Its people "feed themselves with the flesh and milk of camels and with baked cakes". It produces ambergris, ivory, rhinoceros horn, myrrh, liquid storax and "tortoise shell of extraordinary thickness" (same source, p. 128).

The translators read Pi-p'a-lo, Cantonese Pat-pa-lo, as "Berbera, the Somali coast generally", and suggest that the four cities were probably Berbera, Zeila, Mogadishu and possibly Barawa (same source, pp. 128–129). See Chinese records of the Somali coast.

The "land of Berbera" in the Arab geographers

Arab writers used Barbara for a coast and a people, not only for one town:

• Al-Mas'udi, 956, calls the Gulf of Aden al-Khalij al-Barbari, "the Berber gulf" (Fauvelle-Aymar et al., Espaces musulmans de la Corne de l'Afrique au Moyen Âge, 2011, ch. 2, §56).
• Al-Idrisi, 1154, describes "the land of Berbera… where one finds a great number of villages", which begins at a place called Djouah and runs round the Horn as far as Merca and a town called el-Nedja, "the last land dependent on Berbera". "The inhabitants of Berbera feed largely on the flesh of sea turtles" (Jaubert, Géographie d'Édrisi, vol. 1, 1836, pp. 40–45, translated here).
• Ibn Sa'id, 13th century, calls Berbera "the capital" of the Berber towns on the shore (Guillain, Documents sur l'histoire… de l'Afrique orientale, part 1, 1856, pp. 238–239). See Early Mogadishu, Merca and Barawa.

Ruins and graves around the town

• Burton, 1855. Richard Burton visited ruins about a mile south-south-east of the town, including "a small building about eight yards square divided into two compartments", which he took to be a mosque. At the springs behind the town, "the rocks… were covered with ruins of mosques and houses", with "fine mortar mixed with coarse gravel, bits of glass bottles and blue glazed pottery". He also passed "extensive cemeteries" which, he was told, "belonged to a race preceding their own" (Burton, First Footsteps in East Africa, 1856, Project Gutenberg ebook 6886).
• The shore cemetery, 2001. A French team found a cemetery along the shore about 3 km long and 20 to 50 metres wide, with "probably several thousand" graves, some marked by a single standing stone. They suggest it "could have been used continuously from the Middle Ages until today" and note that the sea is wearing away the graves nearest the shore (Fauvelle-Aymar et al. 2011, §§36–37).
• The hinterland. Inland, the medieval town of Fardowsa near Sheekh, about 60 km south, served the caravans going to and from Berbera (de Torres Rodríguez et al., "Towns in a sea of nomads", 2022, p. 265). See Ruined towns of the Somali interior.

A seasonal port

De Torres Rodríguez and his colleagues describe trade on this coast as seasonal. The north-east monsoon made sailing easy between about October and April, when herders also came back to the coast and its wells; in summer the coastal plain "remains almost empty". They suggest that this is why the Portuguese fleet that attacked Berbera in 1518 found it "completely empty of people and goods" (de Torres Rodríguez et al. 2022, pp. 255–257). Later travellers, among them Burton, described the same seasonal fair at Berbera (same source).

Related pages

• Before 1300: the ancient Somali lands
• Early Zeila and the Qiblatayn mosque
• Mogadishu, Zeila, Berbera and Barawa, 1331–1586
• The trade fairs of the Berbera coast, about 1000–1600
• Zeila, Berbera and Mogadishu in the records of Rasulid Yemen, 1229–1454
• Zeila and Berbera in João de Barros's Décadas da Ásia, 1513–1518