Harar becomes the capital of Adal: 1520 to the walls of Harar and Zeila

In short

Today the walled old town of Harar is often said to date back to the 10th or 13th century. The historian Amélie Chekroun, in "Harar as the capital city of the Barr Saʿd ad-Dīn (first half of the 16th century): from its emergence to its fortification", Annales d'Éthiopie 34 (2022): 23–44 (HAL open copy), argues from the written sources that Harar became an important town only when it became the sultanate's capital in 1520. Page numbers below refer to the journal pages.

The founding legends

• Emir Nur. In 1854 Richard Burton was told that the city in its present form was the work of Emir Nur, and that before him there was "a mere mass of Bedouin villages". In the 1920s the Capuchin missionary Azaïs was told that Nur built the ramparts, finished only about 1580 (pp. 24–25).
• Abadir. Chekroun writes that from the late 19th century the oral tradition moved the founding back to the 12th–13th or even 10th century, centred on the saint Abadir. The Arabic Fatḥ madīnat Harar tells of Abadir's arrival from Mecca in 1216 with 405 companions, but it mentions Portuguese and Italians and ends with a prophecy of the Italian conquest of Ethiopia. She cites scholars who date its spread or writing to the late 19th or early 20th century (pp. 25–26).
• Lists of emirs. Some lists of the emirs of Harar start in the late 9th century and join unrelated dynasties, such as the 14th-century rulers of Ifat and the emirs descended from ʿAli ibn Daʾud in the 17th–19th centuries (p. 25).

What the sources and digs say

• First mention. The name "Harar" first appears in the Ge'ez account of King Amda Seyon's wars, which Chekroun, following Bertrand Hirsch, dates to the 15th century. In a list of about 200 territories in a Muslim coalition, Harar supplied only 3 governors, against 99 for Adal and 1,000 for Dakar (pp. 26–27).
• Adal and the Barr Saʿd ad-Din. Chekroun writes that the sultanate's own Arabic sources always call it the Barr Saʿd ad-Din, "the land of Saʿd ad-Din", and that "Adal" was only one territory within it (p. 27).
• Insoll's dating. In 2014 and 2018 the archaeologist Timothy Insoll made the first test excavations inside Harar's walls and dated six mosques that tradition calls the oldest. He concluded that they "post-date the late 15th century" and that Harar was "a late, possibly 16th century foundation" (pp. 28–29). An inscription on the minbar of the Jami mosque, which tradition dates to 1216, gives 1761–1762 (p. 29).
• Why Harar? Chekroun cites 19th-century observers on its cool climate, gardens, springs and position on the caravan road from the Red Sea to the highlands (pp. 29–30).

The crisis before the move

• The sons of Saʿd ad-Din. After Sultan Saʿd ad-Din died at Zeila in 1408 or 1409, his sons took refuge with the Rasulid sultan of Yemen. They came back not through Zeila, then under Christian rule, but through the port of Siyara east of Berbera, and founded the new sultanate in the 1410s–1420s, named after their father (pp. 30–31). For Siyara, see The trade fairs of the Berbera coast.
• Sultan Muhammad (1488 to about 1517). His reign was the longest of the dynasty. In 1517 he and Emir Mahfuz raided the Christian kingdom; Mahfuz was killed, the sultan fled, and in the same year the Portuguese commander Lopo Soares burned Zeila. Chekroun corrects the date of 1514 given by Álvares, because Soares was in Portugal until April 1515 (pp. 31–32).
• Five rulers in a few years. The Futūḥ al-Ḥabasha says Muhammad was murdered on his return, and lists five rulers who followed, each killing or overthrowing the one before: Muhammad ibn Abi Bakr ibn Mahfuz; Ibrahim ibn Ahmad, ruler of Hubat "of the tribe of Balaw"; Wasani; Mansur ibn Mahfuz; and Garad Abun, who ruled seven years until Sultan Abu Bakr, Muhammad's son, killed him. None was of Saʿd ad-Din's line. The official dynastic annals, the Taʾrīkh al-Walasmaʿ, leave them out (pp. 32–33).

The move, July–August 1520

Both Harar chronicles date the move from Dakar to Harar to Shaʿban 926 AH, which Chekroun gives as 17 July to 14 August 1520. The Taʾrīkh al-Walasmaʿ credits Sultan Abu Bakr alone, who "transferred the people and the army from the land of Dakar to Harar". The Taʾrīkh al-mulūk credits him together with Garad Abun, and records Abun's death on 30 May 1525. Chekroun concludes that the move was tied to the internal conflict, but that its exact reason cannot be known. She asks whether it was flight from opponents, a fresh start, a victory of the opposition, or a response to other pressures, including from neighbouring Somali groups (pp. 33–34). For the town before Harar, see Dakar, the lost capital of Adal.

The walls of Harar and Zeila

• Harar. Emir Nur took control of the sultanate in 1551–1552 and ruled until 1567–1568; he killed the Ethiopian king Galawdewos in 1559. The Taʾrīkh al-mulūk describes famine and Oromo raids on Sim, Shewa, Nujub, Jidaya, Dakar and Harjaya, after which "the survivors built fortifications and dug trenches". Chekroun places the building of Harar's wall between about 1560 and 1568. The same chronicle describes a gate and ditch at Harar during an Oromo siege in the 1570s (pp. 35–36).
• Zeila. A late 16th-century Arabic chronicle records that the governor of Zeila, Garad Ladu, had walls built by ʿAtiya ibn Muhammad al-Qurashi, a man of Yemeni origin, while Sultan Muhammad ibn Nasir was away fighting in Ethiopia. Chekroun dates this to between 1575 and 1577. She notes Cerulli's remark that the Arabic word used, darb, survives in Somali as derbi, a wall (pp. 36–37). She places it at a time when Oromo raids on the sultanate continued (pp. 37–38). In an earlier survey, Chekroun and Bertrand Hirsch wrote that the Zeila documents published by Cerulli record the wall being built "to protect the town against Somali nomads plundering" (see The end of Adal, 1543–1583).
• Later views. Burton described Zeila's wall in 1854 as a long low brown wall with round towers and five gates, locked at sunset. It no longer stands (p. 38).
• Other sites. French surveys in 2001 noted possible enclosure walls at Zeila, Amud and Sheikh. The StateHorn project, which is digging medieval sites in Somaliland, reports that such sites usually have no defensive walls. Chekroun notes that the Somaliland sites are mostly undated, while Ifat's towns, such as Nora, had strong walls facing the Christian kingdom (pp. 38–39).

Related pages

• Awfat, capital of the Ifat sultanate, and the tombs of the Walashma
• Imam Ahmad and Adal, 1527–1543
• Futuh al-Habasha
• The Harla and Harlaa
• Abasa and Darbiyah Kola: a medieval town
• First Footsteps in East Africa
• The death of Sultan Saʿd ad-Din at Zeila and his sons in Yemen
• Imam Ahmad's son and the end of the war, 1543–1560
• The end of Adal, 1543–1583: Emir Nur, the Ottomans at Zeila and the move to Awsa