- 1 All dates are AD unless otherwise specified.
- 2 Ghandabalū (غندبلو) in Arabic alphabet, Gǝndäbǝlo (ግንደብሎ) in Amharic and Geʿez syllabaries. In this (...)
- 3 Adam Ali Doyyo (a farmer living in Andodé in Ifāt, died in 1998 at the age of 123; interview: Aleyy (...)
- 4 “Melka” in Afan Oromo means a place where shepherds and cowherds bring their herds for drinking wat (...)
1Oral traditions collected in 1993 and 20021 in the Ifāt region in Ethiopia by Ahmed Hassen Omer refer to Gendabelo2 as a big international market during medieval times.3 The informants explained that merchants were coming from abroad, as far as India, and “many other places.” According to them, Gendabelo was a remarkable meeting point for the commerce of Central and Eastern Ethiopia before the 16th century wars between the Muslim army from Harar led by Imām Aḥmad and the Christian king of Ethiopia’s army in the 1530’s-1540’s (about that war, see Chekroun & Hirsch, 2020b; Chekroun, 2023). They unanimously placed the medieval market town on the left bank of the Awash River in today’s Afar region, at the site of Awash Melka Ballo (district of Buri-Modayto) according to the Oromo, and Awash Melka Kuyyu (further from the highland) according to the Afar: the two locations are about 6 km apart.4
2The Ifāt region, on the eastern escarpment of the Ethiopian high plateau, east of the town of Dabra Berhān and west of the Awash River (fig. 1), has been mentioned in sources since the end of the 13th century. It became the heart of the sultanate of the same name (Ifāt/Awfāt), ruled by the sultans of the Walasmaꜥ dynasty. They reigned over one of the main trade routes between the central high plateau of the Horn of Africa controlled by the Christian king of Ethiopia and the international trade of the Islamic world via primarily the port of Zaylaꜥ on the Gulf of Aden (Chekroun & Hirsch, 2020a). The very patchy documentation from the 14th century suggests that the Walasmaꜥ family may have been dependent on the Christian power, which may have led to unrest at the end of the century, culminating in an open conflict led by one branch of the Walasmaꜥ to free themselves from Christian tutelage. In the early years of the 15th century, what remained of the Walasmaꜥ dynasty abandoned the Ifāt region under full Christian domination. They founded a new sultanate a few years later, named Barr Saꜥd ad-Dīn, between Zaylaꜥ on the shores of the Red Sea and the eastern part of the eastern highlands of the Rift Valley, around present-day Harar (Chekroun, 2020). Therefore, from the 15th century onwards, Ifāt became a region under Christian rule, although many Muslim populations continued to live there. Since then, in the Christian system of government, the Ifāt governors have been called “Walasma” (Chekroun, 2023: 85-86; Hassen Omer, 2020: 103). The main breakbulk markets along the caravan routes from the shores of the Red Sea were then located in this former Muslim territory. The domination of these territories allowed the Christian kingdom to control the major trading cities and thereby connect to the trade network of the Muslim world via Zaylaʿ (Fauvelle et al., 2010: 36-37). Long-distance trade, mainly with the Islamic world, remained in the hands of Muslim merchants, both local and from other parts of the Islamic world (cf. Chekroun & Hirsch, 2020b).
Fig. 1 – Map of the Ifāt region
© Chekroun, 2022
3According to medieval documentation, one of the main markets in the Ifāt region during this period was in the town of Gendabelo, which current oral traditions from Ifāt still mention. This toponym disappears from textual documentation after the 16th century. The identification of three unpublished 19th and mid-20th century texts (one in Ajami mixing Amharic, Arabic and Argobba, the others in Amharic), which mention Gendabelo as an important vanished trade place invites us to re-examine the 15th and 16th century documentation. After having edited and translated these three 19th to 20th century texts, and presented the medieval sources, the objective of this article is to propose a precise location of this important Ethiopian trade site of the late medieval period, based on known textual, topographical and archaeological information about the region.
4In 1992, Ahmed Hassen Omer discovered, by chance, three pages of a paper manuscript (fig. 2; and Hassen Omer, 2020: 295). They were used as cornets for sugar, in a small shop of Dabal, a village in the former waradā (district) of Buri-Modayto, now in the Afar National Regional State (fig. 1). Those pages contain a beautiful poem, a nostalgic lament about the fate of the town of Gendabelo (Ghandabalū), once the “market of the world” (yā-lam ġabyā), now overgrown with brambles, in ruins, deserted by those who once came to trade there, forcing the people of the region to travel long distances in search of markets or to turn to agriculture.
Fig. 2a – 1863 Ajami poem about Gendabelo, folio 73 of an unknown manuscript
© Ahmed Hassen Omer, Addis Ababa, private collection.
Fig. 2b – 1863 Ajami poem about Gendabelo, folio 74 of an unknown manuscript
© Ahmed Hassen Omer, Addis Ababa, private collection.
Fig. 2c – 1863 Ajami poem about Gendabelo, folio 75 of an unknown manuscript
© Ahmed Hassen Omer, Addis Ababa, private collection.
5This rhyming poem with 14 couplets closes with a colophon. It mentions the name of its author (Sheikh ʿAlī Tāju ad-Dīn ʿIssa al-Azharī), the Hegira date and place of writing (August 25, 1863, at the village of Wasīso, in Ifāt), and the copyist (sheikh Aḥmad Walī Muḥammad, on October 15, 1874, at Wasīso). There is no certainty, however, that this manuscript dates back to 1874, and that it is not a later copy of the 1874 original. We do not know what the rest of the manuscript contained, or if it actually ended with this text, as the presence of the colophon might suggest: indeed, the colophon may only apply to this poem, or it may apply to the entire manuscript. The pages are numbered 73, 74 and 75, in Arabic script, written by the same hand as the rest of the text, which, as it is worth noting, is quite unique in an Ethiopian text in Arabic. The text is decorated with numerous seals of Solomon (stars), acting as talismans, according to an Islamic mystical tradition quite widespread in Ethiopia. For example, they could be found on medieval Islamic funerary inscriptions from Ifāt and Tigrāy (Bauden, 2011: 287, 297; Dorso & Lagaron, 2023).
6One of the main characteristics of this poem is its language. Indeed, it is written in Arabic script, but the language is Amharic, mixed with both Argobba and Arabic. While it is quite difficult to translate any Ajami text ‒i.e., an Arabic script used for writing another language, which is a common practice in the Horn of Africa (cf. Gori, 2003 and 2007: 744), it is even trickier to translate a text where three different languages (Amharic, Argobba, Arabic) are mixed together, as is the case in this poem. The combination of those languages in a text written in the region of Wasīso is not very surprising. Wasīso is an Argobba village located in Ifāt on the border of different regions: Amhara, Afar and Oromo, and the area is inhabited by Muslims of different origins: Afar, Argobba, Amhara, Oromo. Argobba is an Ethio-Semitic language like Amharic (Voigt, 2003) and is spoken by Muslim Argobba populations. But today, in Wasīso, the Muslim Argobba peasants do not know the Argobba language, and some of them explain that their ancestors were Christian or Muslim peasants, who spoke Amharic, and had come from the high plateau during the 19th century to settle in the fertile lowlands (Fauvelle, Hirsch & Chekroun, 2017: 260-261). Such tangle of populations, identities, cultures and religions explains the language used in this 19th century poem.
7Nowadays, Wasīso is a small Argobba village, difficult to access. During the 19th century, Wasīso was a sizable center of Islamic learning. Several Arabic manuscripts written by Ifāt ulamas during the mid and second half of the 19th century and preserved in private collections in Ifāt mention the role of Wasīso in Islamic teaching (Hassen Omer, 2006: 22; 2020: 291-292). The main centers that attracted ulamas for their teaching activities in the 19th century in Ifāt, i.e., Tchanno, Aleyyu Amba, Khyaramba, Farré and Wasīso are detailed in an anonymous undated text, soberly titled Al-Ḥabasha, written in Arabic on parchment and dedicated in its colophon to the “Argobba ulama-s of Ifāt,” now kept at Buaa Jartii in Ifāt by the family of Sheikh Salāḥ Marḥaba. In a text dated 1261 H./1845 AD, Al-Islam fī al-Ḥabasha, a sheikh from Ifāt, sheikh Yusūf b. Maḥmud, details the Muslim communities as well as the main Qur’anic schools, ulamas, and the great Muslim leaders of Ifāt. In folios 48-51, much information is given about Wasīso: it was one of the key centers of Islamic learning in Ifāt at that time, a center hosting Muslim students from the surrounding area (Dinki in Gedem and Anṭākiya) but also from further places, such as Karaulé in Awsa or Golima in Rayya. The author specifies that the Argobba, especially from Wasīso, made up a large part of the Ethiopian student body attending al-Azhar in Cairo. The importance of al-Azhar in the training of Ethiopian scholars in the 19th century is also attested in another text, bearing the same title Al-Islam fī al-Ḥabasha but dated 1279 H./1862 AD and written by Sheikh Alī b. Sulaymān. This ulama from Ifāt recounts his “journey in search for knowledge” (riḥla fī ṭalab al-ʿilm) in the mid-19th century, to Ethiopia, including Wasīso, and then to al-Azhar in Cairo, via the port of Aden in Yemen and the holy places of the Hijaz (Mecca and Medina). Since at least the late 15th century, the al-Azhar Mosque in Cairo was one of the major centers of Islamic knowledge for Ethiopians (Loiseau, 2019), and it was still very important in the mid-19th century. Moreover, the sheikh of Wasīso who wrote the poem about Gendabelo in 1863 bears the nisba of “al-Azharī” indicating that he, too, stayed and learned at the famous mosque of Cairo.
Tab.1: Edition and translation of the 1863 Ajami poem
8This text is not the only 19th century Ethiopian document we found mentioning the Gendabelo market and its demise. Another one is a single page written in Amharic by an unknown person at Gunnan (Soddo region), dated Tekemt 2, 1869 EC, i.e. October 11, 1876 AD. This text is copied on an undated parchment manuscript, today in possession of a man named Azmatch Bekel Hāyla Māryām, who lives in Butājirā in southern Shawā, west of lake Zway, in the Gurage Zone (fig. 3).
Fig. 3 – Amharic text from 1876 about salt trade and Gendabelo
© photo Ahmed Hassen Omer, 2015. Azmatch Bekel Hāyla Māryām, Butājirā, private collection.
ከ ባሕረ ኤርትራ ጥግ ጀምሮ በትግሬው አገር እስከ ግንደብሎ አንጦኪያ ጨው ተነገደ። ግራኝ ሲያጠፋት ጨው ተዚያ መሔድ ትቶ የኋሊት ጦቢያ ተነገደ። ተ ጦቢያ ሐልዩ አምባ፣ ተያ በኋላ በመላው ክርሰቲያን አገር ተታችም ጂማ ሄደ። ጨው ተወደደ።
በ፪ ወርሃ ጥቅምት በ፲፰፮፱
ጉነን ተጻፈ።
The salt trade extended from the coast of the Eritrean sea [i.e., the Red Sea] in the country of Tigray to Gendabelo (ግንደብሎ) and ʾAnṭokiyā (አንጦኪያ). Once Grāñ (ግራኝ) destroyed it, the trade route turned towards Ṭobiyā (ጦቢያ) then towards Ḥaleyu ʾAmbā (ሐልዩ አምባ). The salt trade then spread to almost all Christian inhabited areas, even as far as Jimmā (ጂማ). That is the reason why salt became expensive.
Written on the 2nd of Tekemt 1869 E.C.
At Gunnan (ጉነን)
9According to this Amharic text, before the 1530’s, two towns in the Christian region of Ifāt, Gendabelo and Anṭākiya, were the southern ends of the salt route that connected the shores of the Red Sea, probably to Massawa and the Dahlak Islands, to the Christian kingdom. Those towns were destroyed in the 1530’s during the war between Christians and Muslims led by “Grāñ,” a derogatory nickname given by the Christians to Imām Aḥmad (cf. Chekroun, 2023: 220-224). Another Ifāt town, Ṭūbiya, which also appears in medieval documentation, takes on the role of final stop on this route. Probably in the 19th century, it was in turn replaced by the large 19th century Ifāt market, Aleyyu Amba (Hassen Omer, 2020). Lastly, in the 19th century, the salt trade spread to the entire region (about Ethiopian salt routes see Smidt, 2010).
10The memory linked to the greatness of Gendabelo did not vanish with the upheavals of the late 19th and early 20th centuries, including the creation of contemporary Ethiopia by Menelik II or the Italian occupation of 1936-1941. The oral traditions recorded in the late 20th century in Ifāt bear witness to this. During a research project on slave trade routes in Ifāt in March 2022, Ahmed Hassen Omer and Getnet Getu identified an Amharic letter dated tāḫeśāś 13, 1934 E.C./December 22, 1941 AD. This letter is in the private family collections of the late Al-Hajj Muḥammad Daūd Nuré at Khayr Ambā in Ifāt.
Fig. 4. – Amharic letter dated 1941 about trade and the memory of Gendabelo
© photo Getnet Getu, 2022, source: private family collection of the late Al-Hajj Muḥammad Daūd Nuré at Khayr Ambā in Ifāt
አደራ ይስጡልኝ ለሸህ መሐመድ
13/ ታኅሣሥ 34/ ዓ.ም
ሼህ መሐመድ፣ አሰላሙ አለይኩም ይማሙ ሙሔ ነኝ፡፡ ተተገናኘን ቆየን፡፡ አምስት ዓመት፡፡ እነሆ ጥሊያንም ወጣ፡፡ ጠሃይ ነው ተንግዲህ፡፡ ጣሊያን ገበያ ሣይተህል ነበር የኬደው፡፡ ድሮስ ማን ተኸለ፡፡ ሐዲስ አበባም ቢሆን የኋላ ነው፡፡ ባቲም ኋላ ነው፡፡ ትላልቅም አዶሉም፡፡ ታላቅማ ድሮ ቀረ፡፡ አባቶች የነገ ሩንማ ታላቁ ግንድብሎ ነበር አሉ፡፡ ያ አሁን የለም፡፡ አውሣ ልነግድ ፈለግሁ፡፡ የት ልሸጠው ነው ብዬ አሰብሁ፡፡ ባቲ፣ ኤዶመዲኔ፣ ረንዶዬ ኸ አደም ዓሊ ልጆች መንደር፣ አደራ ባራክአሏህ ያስረዱኝ፡ ይምከሩኝም፡፡
ይማሙ ሙሔ ነኝ፡፡
ተቀርሳ ሆኜ የጣፍኩት፡፡
መጪው ረመዳን እንገናኝ ይሁን፡፡
Please let this [letter] reach Sheikh Muḥammad
Tāḫeśāś 13, 1934 E.C.
- 5 Located some 30 kilometers southwest of Kemisé town, north of Ifāt.
- 6 An ancient town not far from Eddo-Mediné to the west of it.
- 7 About Adem Ali, see Hassen Omer,1994: 50-53.
Greetings (āsélma ālékum) Sheikh Muḥammad. This is Yimāmu Muhé. It has been a long time since we met. Already five years. Look the Italians left. The sun is rising [freedom prevailing] as of now. Italy left us without founding a marketplace. Even before, who did so? Even Addis Ababa emerged much later. Bāti also came much later. They are not even big ones. Gone are the days of the big one. Those old (big markets) passed away long ago. What our ancestors told us was that the bigger one was Gendeblo (ግንድብሎ) and the smaller was Hanṭokiya (ሀንጦኪያ). People told us that the cotton market was Gendeblo. This does not exist now. I want now to trade with Awsa. I have a problem finding where to sell [my businesses items]. At Bāti? At Eddo-Mediné?5 at Randoyyé,6 a settlement of Adem Ali’s descendants’ village?7 Please Allah bless you for your brief suggestions and pieces of advice too.
I am Yimāmu Muhé.
Let us meet in the coming month of Ramaḍān.
- 8 A small village town located between the towns of Bāti and Degan, along the route from Kombolcha to (...)
I wrote it when I was at Qarsā.8
11It was written by a certain Yimāmu Muhé and addressed to a certain Sheikh Muḥammad, both otherwise unknown. It was the end of 1941: the Italians had just been driven out of the country after their five-year occupation. Their presence had disrupted many things in the country, including economic matters: the author laments, for example, that it is now impossible to find a decent market in Ifāt, and that one has to go to remote areas, such as South Wallo. Gendabelo, the big market in Ifāt, has not been in existence for a long time, so he has to turn to other places. It is interesting to note the geographical similarity between Gendabelo, on the escarpment between the highlands and the lowlands, a breakbulk point for caravans connecting the Gulf of Aden to the medieval Christian kingdom, and the town of Bāti mentioned by Yimāmu Muhe as a replacement. The market of Bāti is indeed:
an important market located 65 km east of Däse on the Komboltcha-ʿAsab road, at the base of the Rift Ralley escarpment. [..] Because of its location, Bāti serves as a major cultural bridge between the highlanders, mostly Christians, and the lowlanders, largely Muslims. Historically, Bāti has always served as a staging ground for the camel caravan between the coast, mainly the ʿAsab area, and the lowland areas of Tégray and Wallo. (Wubneh & Morin, 2003)
12This position between highlands and lowlands, a little further north on the escarpment, thus remains strategic for the location of large markets, even several centuries apart. And the past greatness of Gendabelo, and to a lesser extent that of Anṭākya, seems to be more important than these new markets, between which the author can’t choose.
- 9 As Crawford points out, the name of this river does not appear in Cecchi’s 1883 work, which Castro (...)
13The market of Gendabelo disappeared but the memory of its grandeur was still vivid when these texts were written in the second half of the 19th century and middle of the 20th century and it seems that the toponym survived in the area. As mentioned by O. Crawford in his commentary on Zorzi’s itineraries (Crawford, 1958: 96; quoted partially by Huntingford, 1989: 107), the late 19th century European maps of the area note that the southern branch of the stream which originates at Aleyyu Amba bears a name similar to the 15th- and 16th- century Gendabelo. On an Italian map reproduced by Lincoln de Castro in 1915, the stream bears the name “Kindl Bell”9 (Castro, 1915: vol. 2, opp. 240). On the de Chaurand’s map published in 1894, the main stream is called “Ghinde-vel” (Chaurand, 1895: 60 and 1894: “Antotto Jg”). Similarly, as mentioned in our introduction, both Afar and Oromo late 20th century oral tradition place Gendabelo near the ford of Melka Kuyu on the Awash, in the very area where the Ajami poem was found.
14In medieval documentation, the town of Gendabelo appears as an important stop on the routes between the Christian kingdom of Ethiopia and the shores of the Red Sea in the 15th and 16th centuries, as well as a large market run by Muslims. To our knowledge, the first mention of the city of Gendabelo is in the chronicle written in Geʿez of the reign of the Christian king Ba’eda Māryām (r. 1468-1478). In the early part of his reign, the king led an expedition to the eastern escarpment of the Ethiopian highlands, into Muslim regions, after befriending the Sultan of the Barr Saꜥd ad-Dīn (the “Adal” in the Geʿez text). From Tigrāy, where he had settled, he led his troops against the Dobꜥā, a Muslim population in south Tigrāy, conquered the country, and established troops and a church there. On his return from this victorious campaign, the king came to the region of Gedem, then to Lāwo Gabayā. Then he went to the country of Qatchéno in the region of Ifāt. Having stopped there, he inquired about the names of the cities (āgar) in the region:
He was told, “This one is Makré, a place of relegation, and this one is Gend Belo (ግንድ፡ ብሎ), where we are and which is (also) called Falaga ʿAgāt, in the land of Gādāwi.” He then inquired whether the people of Adal came to this area and was told, “Yes, they come and spend the day there.” [...]. The king then came to the region of Ṭobya in Ifāt and pitched the royal woolen tent there – in the past our king Dawit had lived in this country – and developed many plantations there. Then he went out from there through the Dago Mācheg pass and came to Falaga Gālāy (or Gāsāy), which is the country of Dabra Berhān. (Anonym, Zar’a Ya’eqôb et Ba’ed Mâryâm: 152; with the kind help of the late Dr Berhanou Abebe (1933-2008) and corrections by Schneider, 1990: 129)
15King Ba’eda Māryām thus descended into the Ifāt region from the north (the Christian region of Gedem) and followed a north-south route that roughly ran parallel to the escarpment before ascending to the highlands of Dabra Berhān through the Dago Mātcheg (i.e., “Dago’s fort”) gorge.
16This space also appears in an itinerary handed down by the Venetian scholar Alessandro Zorzi, who, in the early 16th century, collected geographical information from Ethiopian monks and pilgrims visiting Venice, and described routes to, from and within Ethiopia. In 1523 in Venice, Zorzi interviewed an Ethiopian monk, named Anṭonyos, originally from Bugna (“Urvuar”), on his way to Portugal after a journey to India. Among all the itineraries collected by A. Zorzi, this is the only one detailing an oriental route (east-west) from the port of Zaylaꜥ to the royal camp of the Christian king Lebna Dengel, at Barārā in South Shawā:
He says that the fleets of the Moors from Combaia bring much merchandise – spices and cloth of gold and silk – to the Red Sea, to the port of Zelo, and thence on camels through the province of Adel, that is of a Moor of Adel, to the city of Genasere of a Moor, which is 25 days; and the region is desert and pasture for animals. And from Genasere one goes to the city of Gendevelu through cultivated country, crossing more provinces, and it takes 40 days. This Gendevelu is a great mercantile city, and it is of the Presta Davit, where the said caravans of camels unload their merchandise in warehouses; and it is the merchant ships (“forcio mercatanti”) of Combaia that bring all the spices except ginger, which is found in this land of the Presta. Then that merchandise is obtained by purchase or barter by the merchants of the Presta, and the currency is Hungarian and Venetian ducats, and the silver coins of the Moors, and by that route various things are brought from the whole of India; and so too on the return journey these of the Presta go to the said city of Zelo, that is an excellent port in the Red Sea, on the Aden route, that is in Arabia Felix. And the said city of Zelo, that is in Aphrica, is of the Moorish king of the province of Adel. And also, he is king of Genasere city, and of the cape of Guardafumi and of Barbara. […] From Zelo one goes westwards to Genedevelu city, and from there southwards to Barara. (Zorzi, Itineraries: 172-175)
17This “itinerary has been added later at the bottom of the page and continued in the left margin” (Crawford, 1958: 172 note 1):
And from Gendevelu to Tobia city, 1 day
From Tobia to Zarma, 2 days
From Zarma to Gogo, 1 day
From Gogo to Debrebraan, 1 day
From Debrebraan to Zancar, 2 days
From Zancar to Barara, 1 ½ days (Zorzi, Itineraries: 172-173)
18This itinerary gives precious information on Gendabelo (“Gendevelu”); this city appears as the great trading city, the breakbulk port for the caravans of the trade between the Muslim lowlands and the Christian highlands (“by this road, various things are brought from all over India”). Anṭonyos thus specifies that the goods (spices, precious fabrics) came from Cambay (“Combaia”), in Gujarat, India, transported by boat by Indian Muslim merchants, to the port of Zaylaꜥ (“Zelo”). These goods were then loaded onto camels and followed the caravan route through the sultanate of Barr Saꜥd al-Dīn. On this itinerary, the sultanate is named “Adal,” as was then customary among Ethiopian Christians (Chekroun, 2020: 132-133). But there is no doubt that it was the sultanate which controlled, as Anṭonyos explains, the port of Zaylaꜥ and the port of Berbera, as well as the large city at a 25 days’ walking distance inland that he calls “Genasere”. This city is not identified today, but it probably stood at the eastern end of the eastern high plateau of the Rift, in the vicinity of the present-day city of Harar. The area is indeed “desert and pasture for animals.” The second segment, from Genasere to Gendabelo, the longest part of the journey (40 days) probably went through the Tchertcher (“cultivated country”) to Ifāt. Here we are in a better-known country, with denser settlements. It is identified as Ifāt by the presence of the toponyms Ṭobya and Gendabelo, known as Ifāt cities, and not far from Dabra Berhān, a large Christian city that dominates the escarpment where the Ifāt region is located. About Gendabelo, the itinerary tells us that it was under the rule of the Christian king, “Presta Davit”, Lebna Dengel (r. 1508-1540) whose reign name was Dawit. The city was home to warehouses for goods unloaded by Muslim merchants, which were sold to Christian merchants, who then took the goods to the Christian highlands. Exchanges were made “by purchase or barter”, and several currencies were used: Hungarian and Venetian ducats, and mostly silver dirhams from the Islamic world (Chekroun & Hirsch, 2020a: 110-111). It was therefore a very important trading place, a point of contact between the Christian kingdom and the long-distance trade conducted by Muslim merchants.
19As Anṭonyos explains, the central breakbulk role of Gendabelo was due to its strategic position on the escarpment of the Christian high plateau. Camel caravans, which could carry heavy loads from the coast could not continue their route beyond the Ifāt region. Climbing and living on the high plateau (which rises to 2000-3000 meters above sea level) is not possible for camels. The unloading of camel caravans and loading onto other beasts of burden, especially donkeys and human backs, therefore required important infrastructure: the warehouses mentioned by Anṭonyos. Gendabelo’s role as a breakbulk point is confirmed in a slightly later source endogenous to the Islamized territories of the Horn of Africa:
The Imām had sent Warajār Abūn to Zaylaꜥ to buy cannons for him so that he could take this fort [i.e. Ambā Geshen]. He bought a large bronze cannon, and two little ones made of iron and carried them by camel as far as the city of Ghandabalū. […] Abbas then had them carried on men’s backs, because the camels could not traverse the path. (Shihāb ad-Dīn, Futūḥ: 344)
20This quotation is from the Futūḥ al-Ḥabasha, a lengthy account in Arabic written by a Yemeni nicknamed Arab Faqīh in the mid-16th century, which details the various attacks on Christian territories bordering Muslim territories by the armies of the Barr Saꜥd al-Dīn Sultanate led by Imām Aḥmad in the 1520s, and then the conquest of almost all of the Christian kingdom’s territories by the same armies in the 1530s (cf. Chekroun, 2023). This excerpt on the purchase of firearms is not the only one mentioning Gendabelo. Much earlier in the account, during Imām Aḥmad’s first visit to Ifāt, Arab Faqīh devotes several lines to this city and the army’s stay there:
The Imām pitched a white tent in the area of Anṭūkyā which was part of the territory of Ifāt in the land of Ethiopia (ḥabasha). […] On the fourth day the imām put his forces in order and set out for the city of Ghandabalū […] They marched all that day along a narrow route up hill and down dale, with men and animals becoming exhausted; and on that day, on account of the narrowness of the path, men had to carry the imām’s wife on their shoulders. At sunset the tent of the imām Aḥmad was pitched in a place called Daq, where qāt grew in abundance. […] In the morning the army set out for a place called Bāzmalī in the district of Ifāt, in the country of the infidels. It was a forbidding place. They spent the night there and next morning, they went on to Ghandabalū in Ethiopia. It was ruled by the king of Ethiopia but had Muslims living in it who paid the jizīa (tax or tribute) to the king. The people of Ghandabalū, the sheikhs, the faqīh-s, the merchants, the people of the villages, came to the imām Aḥmad and his army, and honored them. The Muslims helped them with twenty ounces of gold. […] The imām intercepted some infidel traders in Ghandabalū who had in their possession valuables belonging to the king of Ethiopia. He killed them in the middle of the town, and took their valuables, their mules, and their hacks, along with their loads. […] He set out from there in the afternoon to return to the country of the Muslims taking with him the treasure of the king of Ethiopia. They spent that night close to the Awāsh. (Shihāb ad-Dīn, Futūḥ: 36-39; Shihāb ad-Dīn, Conquête: 28)
- 10 About Medieval towns in the Christian kingdom, under the authority of a Christian ruler, but inhabi (...)
21As a center of long-distance trade, Gendabelo was very wealthy: the Muslims were able to offer a very large amount of gold to the imām (with the “twenty ounces” the imām had “one-hundred swords” purchased) and the Christians also seemed to be in possession of many goods. Gendabelo appeared as a city under Christian authority, inhabited by Christian merchants, but mostly by Muslims; not only merchants from the coast, but also people who resided there permanently: peasants, and also men of religion, scholars, and probably merchants running the warehouses and serving as intermediaries between passing Muslim merchants and Christian merchants who came to trade. A very large Muslim community thus lived in this city under the authority of the Christian king.10 These Muslims, especially the merchants of Gendabelo, also rallied the camp of Imām Aḥmad to fight the Christian king when the army decided to move towards the north of the kingdom:
Those people of Ifāt and Ṭūbyā who had converted to Islam, along with the people of Ghandabalū, and its merchants, came and gathered in a group […] Uraʾī Abūn set out to join the imām, and the people of Ifāt with their cavalry, around fifty knights and two-thousand infantry, with their wives and children, went with him. They marched on and reached the imām who was on the way to the land of Dawāro. (Shihāb ad-Dīn, Futūḥ: 296)
22Thus, the precious imported merchandise and merchants of Gendabelo, whose disappearance is lamented in 19th century texts, are well attested by 15th- and 16th- century sources. The very last mention in the early sources is less explicit about the city’s commercial role. In an anonymous, untitled Arabic text from the very end of the 16th century, produced within the Islamized territories of the eastern Horn of Africa (Harar, Awsa), an individual bears a nisba (name showing an affiliation) that may be linked to Gendabelo (Cerulli, 1931: 72). The text, which Héloïse Mercier calls Wa kāna ‘alā khizānihi, deals with the history from 1576 to 1591 of the Awsa region, an oasis region close to the present-day border between Ethiopia and Djibouti (Mercier, 2020: 117-140). The text begins with the reign of Imām Muḥammad b. Ibrāhīm Jāsa (1576-1583), who was the one who transferred his power from Harar, the main city in the eastern Islamized territories since 1520, to Awsa in 1576. The text begins with those words: “His treasurer was Z.ḥ.l b. ꜥAbd Allāh al-Ḥabashī and his secretary was the kabīr Ḥāmid b. Ibrahīm al-J.n.d.b.lī (الجندبلي).” (Mercier, 2020: 129 and 136) Thus, at the beginning of his reign, Imām Muḥammad had as his main advisors a man from the territories of the Christian kingdom (al-Ḥabashī) and a man connected with the city of Gendabelo (al-Jandabalī). The city was therefore important enough for someone to have its nisba. This also indicates that at the very end of the 16th century it was still a current nisba. Therefore, the place probably still had a certain status, at least it still echoed in the minds of the men from the Islamic territories who used to trade in the Gendabelo market. To our knowledge, however, this is the last occurrence of the mention of Gendabelo before the sources of the 19th century and their memory of the past grandeur of the city.
- 11 See “Ethiopia - Quest For a Lost Muslim City”, All Africa, July 29, 2009: online: [https://allafric (...)
23Even though the discovery of the archaeological site of Gendabelo was announced in the press,11 the precise location of the medieval place is still unidentified. With the valuable help of the poem in Ajami, it is possible to define a relatively small area, which coincides with the archaeological sites of ruined medieval Muslim cities identified in 2007 in Ifāt: Asbari, Messāl, Norā, and Beri-Awfāt (Fauvelle-Aymar et al., 2006; Fauvelle, Hirsch & Chekroun, 2017). Given the number of sites in an extremely small area, and the number of important towns mentioned in Ifāt for the 15th and 16th centuries, it is –and probably always will be– impossible to identify with certainty which of them is Gendabelo. Medieval sources also describe Ṭūbya and Anṭākya as important trading places. Unless one finds an inscription explicitly mentioning the name of the city, it will be impossible to be sure.
24The Ajami text gives valuable information on the region where the market town of Gendabelo was located by naming inhabited places associated with it: Wasīso, Rāssā Gubbā, Tchanno, Wārkāmbā. These are all places located on the left bank of the Awāsh, within a rather limited perimeter. The text also lists the populations that inhabited this area in the 19th century (and that can easily be found today in this area): the Afar (the “Adal” in the text), the Oromo (the “Galla” in the text), the Shājurā, the Qātchino. The author suggests a series of geographical positions, which, when projected onto a map, are easily visible by an observer looking from the Wasīso/Rāssā Gubā region to the South/Southwest, and classifying the places or populations to his left and right, i.e., to the West and East:
25Wasīso (plain) – Rāssā Gubā (hill)
26Afar (in the plain) – Oromo (on the high grounds)
27Shājurā de Tchanno – Wārkāmbā
28Qātcheno (to the West) – Amhārā (to the East)
29Of course, this distribution of the area’s populations is contemporary with the author and not with Gendabelo’s flourishing period in the 15th and early 16th centuries, before its disappearance following the conquests of Imām Aḥmad in the 1530s. It does however reveal an axis, which, drawn on a map, coincides with the 19th-century trade route leading from the Ankober narrow pass to the Wasīso region, between Wārkāmbā and Tchanno (Hassen Omer, 2020: 187ss). It is quite likely that Gendabelo was located on this axis, perhaps a little further north, in the area of Asbari, which is in the district of Rāssā, known today as Awash Malkaa Bello by the Oromos and Malka Kuyyu by the Afar.
30The topographic map made by the Ethiopian Mapping Authority in 1992 “0940A1 Abaye Atir 1:50,000” shows the small mountain range mentioned by our author, which represents the last high point between a plain to the east with an altitude of about 1300-1350 meters before the highlands (2000 m high just a few kilometers to the east) and the lowlands to the west in the direction of the Awash, whose altitude decreases rapidly to below 700 meters. “Rāssā Gubā” (“Rasa Hill”) appears on this map, with a maximum altitude of 1438 meters, while Wasīso, the village at the foot of the hill, is named “Kedeburā”. During our various visits to this area in the last 20 years, this village, inhabited by Muslims calling themselves Argobba, was indeed named Wasīso (Fauvelle, Hirsch & Chekroun, 2017; Hassen Omer, 2020).
Fig. 5. – Main mosque of Asbari
© Photo: Nora/Gendebelo Program 2009
- 12 The site of Asbari seems to be mentioned for the first time in 1893 by L. Traversi (1893: 683) and (...)
31For the moment, and in the absence of proper excavation of these sites, Asbari12 seems to be the site which, best coincides with the image of Gendabelo given by the 15th- and 16th- century sources. Indeed, it is the site that displays the most monumental ancient mosque on Ethiopian territory. It is a very large rectangular-shaped building (approximately 29 m by 28 m); its walls are an average of 4 meters high and one meter wide (fig. 5; Chernet, 1990: 308; Fauvelle-Aymar et al., 2008b: 139-143; Torres Rodríguez & Cornax Gómez, 2023). The urban site extends over 40 hectares, preserves traces of an imposing enclosure wall, two other mosques and a large cemetery. A thorough archaeological study, including a complete survey and mapping of the remains and excavations, would be very valuable.
Fig. 6 – Some representative finds on the surface of Asbari (Ifāt).
1. Ostrich eggshells; 2. Worked and un worked marine shell and shell fragments; 3. Blue and white Chinese porcelain; 4. Drawn tinted glass and stole beads, including polychrome glass bangle fragment; 5. Blown glass; 6. Slip-painted glazed ware; 7. Metal.
© Nora/Gendebelo Program, Dec. 2009, A. Chekroun 2022
32During a very brief two-day visit in December 2009, a very important and rich material was collected on the surface (fig. 6). The results of this brief visit have not yet been published, but we show here some representative finds which testify to the richness of the place. Worked and unworked marine shells and shell fragments were identified. As Timothy Insoll demonstrated, those found in the trade center of Harlaa (Eastern Ethiopia) between the eleventh and early fifteenth centuries were mostly imported from the Red Sea and the Gulf of Aden where they are indigenous (Insoll, 2021), but they may also have come from the Indian Ocean. Further study is needed, but it is possible as of now to identify cowries (cypraea), olivia bulbosa and engina mendicaria, as well as cypraea dorsa and some with dorsum removed (fig. 6.2). Numerous colored, round and elongated spun glass beads, including some twisted, were also collected (fig. 6.4). While some are probably local, most of them were most likely imported. The red and yellow fragment of polychrome glass bangles may be related to those that A. Gonzalez-Ruibal dates back to the 15th to 17th centuries on coastal sites in Somaliland, and points to places of production in Yemen, and also in Egypt, Syria, Iran and India (Gonzalez-Ruibal et al., 2021: 15). Numerous shards of local ceramics, some with designs (lines, holes, notches, chevrons...), will have to be examined more closely, in relation with other better known Ethiopian collections (cf. Taits, 2020). Several shards of imported ceramics, including blue and white Chinese ceramics (fig. 6.3; cf. Parsons-Morgan, forthcoming) and Islamic ceramics (fig. 6.6), some of which were reworked, clearly point to prestigious long-distance imports. Other elements also testify to the wealth and importance of the city: metal rods (fig. 6.7), ostrich eggshells, some with dot designs (fig. 6.1), fragments of polished glass (fig. 6.5); obsidian tools... So many elements collected in a few hours, which clearly indicate a strong link with long-distance trade and a concentration of luxury objects coming from local production as well as distant imports (from the Islamic world, the Red Sea, the Indian Ocean). In comparison, the site of Nora, less than fifteen kilometers south of Asbari, which was partially excavated during a one-month mission in 2008, revealed much less artifacts of this kind, and even if the mean occupation date was clearly related to the 14th-century Walasmaʿ dynasty (cf. Fauvelle, Hirsch & Chekroun, 2017), the site of Beri-Awfāt revealed some artifacts from the 15th and 16th centuries (such as monochrome Chinese blue ceramics) but in far less quantity and with far less diversity than the artifacts collected at Asbari.
33In our opinion, the Asbari site represents a good clue to help identify the location of the 15th-and 16th- century Gendabelo. As mentioned earlier, there are other itineraries which also provide us with elements for the location of the city. In the itinerary reported by the monk Anṭonyos to Alessandro Zorzi, the most detailed portion is that which goes from “Debrebraan” (Dabra Berhān) to “Gendevelu” (Gendabelo), with three stages of one- or two-days’ walk, which testifies to the fact that the traveler is on well-known ground, i.e., in an area under the firm control of the Christian kingdom. The itinerary’s starting point is the only detail known to us: Dabra Berhān, a royal camp and city founded by King Zarʾa Yaʿeqob at the end of his reign, around 1454. The city of Gendabelo is located at 5 days’ walk. A careful examination of the topography of the escarpment in this region and of the passes allowing access to the lowlands pleads in favor of a continuity of routes for a long period of time. The passes going down from the high plateau to the lower lands are few and the only one likely to have been used is located at one day’s walk from Dabra Berhān and was still in activity in the 19th century to go to Aleyyu Amba from Ankobar and it is more than likely that it was in use at an earlier time. The toponym Gogo mentioned on the route is therefore likely to refer to the Dago Metcheg, a pass used by King Baʾeda Māryām on his return from Ifāt to Dabra Berhān. Yet, the toponyms cannot be assigned to specific places and distance is only measured by days of walking, which obviously makes it hard to evaluate. Indeed a day’s walk down or up the escarpment could not equal a day’s walk on the less rugged intermediate plateau, where the cities of Ifāt were located.
- 13 Ṭobyā, a town in Ifāt mentioned in the chronicle of Ba’eda Māryām and located on the itinerary draw (...)
34An itinerary recorded by Antoine d’Abbadie based on information given in 1839 by M. Dufey, a French traveler who had traveled this route, provides useful points of similarity between their conditions of travel, on foot and with pack animals, and those of more ancient times (Abbadie, 1890: 314-315). Leaving Ankobar, the traveler and his companions passed through Aleyyu Amba on the first day, then went up towards the East-North-East to establish their camp at a little more than three hours’ walk of Aleyyu Amba. The next day, they reached Tchanno after a one day’s walk. Then their route turned eastwards, towards the Awash River, through Afar country, past the Awadi River, where they camped on the third day. After four days, including two days of rest and two days of walking, they reached the Awash river. The elapsed time for this itinerary is quite similar to the time spent by A. Zorzi, even if the latter completed the distance from Gogo-Dago Metcheg to Dabra Berhān in one less day. There is one important difference however, and that is that if the descent of the escarpment is comparable, with intermediate stages that are obviously not superimposable (Aleyyu Amba, Tchanno are towns of more recent creation), M. Dufey’s itinerary heads towards Awash, while fifteenth‑century travelers would almost certainly have passed through the towns of Ifāt, those ruined cities to the northeast that we explored during our archaeological research. If we accept such a premise, we can put forward as a strong hypothesis, consistent from the point of view of the known itineraries and the walking time between the stages, a possible identification of Nora with Ṭūbya13 and of Asbari with Gendabelo.