Report – Ancient places in the chronicle of King Galāwdewos
In our first report we declared our aim to annotate place names in the first 50 chapters of the Chronicle of King Galāwdewos. We have managed to placeName elements in the TEI-XML edition of the entire Chronicle, and it was a very interesting task on which we are going to report now.
Mapping marked places in the chronicle and a historical description
To provide a better understanding of the place names which are marked up in the Gǝʿǝz text of the Chronicle of King Galāwdewos (CAe 3122, id: LIT3122Galaw), it is very important to explain the king’s routes, the places he passed through and the places where he settled. The narrative of the story of King Galāwdewos’s movement begins in the north of the Christian kingdom, in Tǝgrāy, where he was crowned in the monastery of Dabra Dāmmo, the only monastery inaccessible to the Muslim forces, which was burnt and destroyed by Turks in 1557, as we will see later. According to the chronicle, the king began to fight the Muslims in the province of Tǝgrāy, one of the strongest holds of the Christian kingdom in the medieval time. However, he was defeated. The king at that time was not strong enough to face his great enemy of the Muslim leader Imām ʾAḥmad b. ʾIbrāhīm al-Ġāzī, who successfully defeated his father King Lǝbna Dǝngǝl (1508–1540) in all battles of 1529–1540. For reasons of military strategy and in order to gather and organize a stronger army, Galāwdewos decided to flee to Šawā, a province where his predecessors were accustomed to camp regularly. Then, the chronicler mentions the king’s route like this
Mār Galāwdewos then crossed beyond two rivers from the direction of Tǝgrāy to the direction of the land of Šawā in order to see there those of his flock in righteousness and his innocent people in equity. He reached the country he wished in the month of Ḥazirān which is the month of Sane, the beginning of winter months of the Abyssinians.
In the medieval times the two big rivers are Takkaze that divides Tǝgrāy from Gondar and the other is the Blue Nile that divides Goǧǧām from Šawā. These two rivers, even if they are not mentioned by the name in the text, are certainly referred to here. An abstract entity to refer to is therefore referred to adding an empty element in the text. In the same passage we have places which are named and are thus marked up with a reference to the relative place entity. Below is the sample of the marked-up places directly taken from the text (see below for a clarification of how we decide which entity to refer to).
ምዕራፍ፡፱። ወእምዝ፡ ዓደወ፡ ማር፡ ገላውዴዎስ፡ ማዕዶተ፡<placeName ref="http://betamasaheft.aai.uni-hamburg.de/LOC6850Takkaze"/> <placeName ref="https://www.wikidata.org/wiki/Q882739"/>, <placeName ref="http://betamasaheft.aai.uni-hamburg.de/LOC1022Abbay" notBefore="1540" notAfter="1541">ክልኤቱ፡ አፍላግ፡</placeName> መንገለ፡ ምድረ፡<placeName ref="http://betamasaheft.aai.uni-hamburg.de/LOC5597Sawa" notBefore="1540" notAfter="1541">ሼዋ፡</placeName> እመንገለ፡ ምድረ፡<placeName ref="http://sws.geonames.org/444187" notBefore="1540" notAfter="1541">ትግራይ፡</placeName> ከመ፡ ይርአይ፡ መርዔቶ፡ እለ፡ ህየ፡ በጽድቅ፡ ወይዋሄ፡ cሕዝቦ፡ በርትዕ። ወበጽሐ፡ ኀበ፡<placeName ref="http://betamasaheft.aai.uni-hamburg.de/LOC5597Sawa" notBefore="1540" notAfter="1541">ምድር፡</placeName>ዘፈቀደ፡ በወርኃ፡ ሐዚራን፡ ዘውእቱ፡ ወርኃ፡ ሰኔ፡ ርእሰ፡ አውራኀ፡ ክረምቶሙ፡ ለእልኅቡስ።
After the king settled in Šawā in the first year of his reign in 1541, he moved for some time from one locality to another recruiting armies from the Christian communities and then finally he stationed in the mountainous region of Ifāt where he fought the then governor of the region Naṣraddin, son of Imām ʾAḥmad. In the first confrontation, Galāwdewos was defeated but in the second confrontation, he became victorious. This victory helped him to attract a large number of Christian adherents who joined him.
The network of places involved in this text is far from being limited to the local history, and we need to mark up entities which are referred to, like the Ottoman empire or Portugal. In facts, during the second year of the reign of king Galāwdewos (1541), the Portuguese soldiers arrived on the coast of the sea to assist the king and sent a messenger to him to join them as soon as possible. When the news reached Galāwdewos; he marched immediately towards the north in similar routes crossing again the Blue Nile and joined the Portuguese. After some preparation, the king confronted the big enemy Imām ʾAḥmad b. ʾIbrāhīm al-Ġāzī in 1543 at a place called Wagarā, which was the strong hold of the imām for almost a decade, north of lake Tānā. ʾAḥmad was killed and his soldiers retreated. This was a turning point in the reign of King Galāwdewos. Following the victory, many muslim forces a long with the wife of ʾAḥmad, according to the chronicler, retreated towards the sea, which means definitely the Red Sea, which is the entity we mark up, although the reference is generically to the sea. However this was not yet a complete victory over the Muslims because there was one of the most powerful war lord of the late Aḥmad by the name called ʾAbbās who was governing the southern Muslims states, the most fertile regions including Dawāro, Faṭagār, and Bāli from 1539–1544 and this situation shifted the king movement again towards the south.
According to the chronicler, the king prepared to confront ʾAbbās, and then he left the place where he camped (he must have been at that moment in around Wagarā) to southwards crossing two rivers and arrived at a place called ʾAgāy, a locality in the region of Waǧ. Here, the two rivers referred in the chronicle are the Blue Nile and ʾAwāš. When ʾAbbās heard the news of the arrival of the king in ʾAgāy, he organised his army from the regions he was governing and decided to confront the king before he expanded to his territories. He came and stationed in the region of Waǧ which was under the king’s domain. The battle took place in October 1544 and ʾAbbās was killed and a large number of his armies were also slained. The king now managed to regain the lost three southern provinces of the Christian kingdom under the reign of his father, King Lǝbna Dǝngǝl. Peace and stability was also relatively maintained in the region. Then, the king established his temporary capital at a locality what was called ʾAgrāro, in the fertile province of Dawāro between 1544–1545. It is evident that in the coming five years ʾAgrāro had been a center for the king royal seat from where he conducted military campaigns to the southern and western of the Christian kingdom. It is then clear how also the administrative structure of the Christian kingdom is not easy to map, with an unsettled capital and frequent changes in power: this is one more reason of interest in the annotation and documentation of attested places and routes in the historiographical works.
The king conducted a campaign to the Haddiyā region, south of the Faṭagār to fight the people of Haddiyā and subsequently, he led several campaigns to the Gāllā people, who were yet on the southern border of the Christian kingdom. This was a period from 1544 to 1550. In this section we would need to mark ethnic names and curate our records to include information about where these population were attested, which we have not done yet, although we already have referrable entities for most groups.
In the eighth year of his reign in 1548, as the chronicler points out, the king launched a new campaign from his center ʾAgrāro to the region of Dāmot, which lay south of Goǧǧām to punish the pagan people. However, he assigned his ablest general Fānuel to lead a campaign in the same time to the strong hold of the Muslims state of ʾAdal, east of the region of Ifāt. Both the king and the general won a victory and collected an abundant booty that they distributed to people in the court.
To consolidate his kingdom, Galāwdewos built a capital city in a locality of Waǧ in 1550–where the king settled for the rest of his life. It was perhaps the first fixed capital of the Solomonic Kingdom during the medieval time. The Chronicle also devoted a chapter to describe how the palace of the king was built by the hand of the Egyptians, Syrians and other foreign populations, which we also mark up in the text as place names. The king from here planned the military campaigns and the reconstruction of churches and the production of religious texts for the revival of the Christian culture. Among these he built the royal church called Tadbāba Māryām (1552).
From his new capital the king undertook an expedition to the kingdom of Gambo, which was located near Dāmot in 1552. This kingdom was named after the name of its people called Gambo whom we do not find today in the ethnic map of Ethiopia.
In the following years 1553-1557, Galāwdewos directed the campaign to the inhabitants of Gumar, a district to the south of Faṭagār who had been in revolt against him for five years. In the same way, he was also engaged in fighting the Gāllā in the province of Bāli.
Though the king was successful in maintaining peace and consolidating his power, he faced another threat from the North of the Christian kingdom that changed the direction of military confrontation once again from south to north. In 1557, the Ottoman Turks penetrated the north of the Christian kingdom stationed at Dǝbārwā, the capital of the Bahra Nagāš and burnt the oldest monastery of the Christian kingdom Dabra Dāmmo where the king was crowned and the king father King Lǝbna Dǝngǝl and other saints buried. At that time the Turks had also an alliance with local chiefs who were loyal to the previous leader of the MuslimʾAḥmad and collaborated against Galāwdewos. The Turks also invaded a hot and desert province of Mazāgā in 1558 but they failed to endure the climatic condition of the region. The Muslim queen by the name called Gāʿǝwā, whom the chronicler wrongly represented as the queen of Salawā, was also on the side of the Turks. However, due to the alliance of the Christians of the Bur, a locality in the province of Tǝgrāy, Galāwdewos defeated the Turks and maintained the region under the autonomy of the Christian kingdom.
On the other hand, the Mulsim of the ʾAdal state again revived and began to invade the southern provinces of the Christian kingdom which the king took over after the death of ʾAbbās. Nur, the niece of the greatʾAḥmad, stationed his army in the region of Faṭagār during lent of 1559. The king decided without proper preparation to confront Nur and he was killed by Nur forces in 1559 in Faṭagār. They cut his head as a war trophy. The chronicle mentions that he was buried in the church named by the name of Galāwdewos, but we are not sure exactly about the location of this church.
This brief account of the story of the king activities in space and time shows some of the main problem in the annotation of places:
- the identification of places generically referred to, as “the north border”, “the sea”, “the mountains”
- the location of named places for which only relative location is provided and no archeological evidence is available
- the identification of the appropriate entity to refer to for each place
- the location of ethnic groups
Identifying and Marking up Ancient place names in the chronicle of King Galāwdewos (1540–1559)
This phase of the project dealt with the ancient places attested in the Chronicle, by identifying and marking up those place names in the Gǝʿǝz text, once this was copied in a raw form in a Textual Unit record of Beta maṣāḥǝft. We aim at mapping the places attestations and provide a historical explanation in relation to the history of the king activities pertinent to the period under discussion. Although for the current visualization of this record we can offer only screenshoots, there are two up-to-date and accessible versions of the TEI source file which you can see.
The preparation of this record, which is being edited to be a full critica edition of the text, required some initial encoding stages and the text is not ready as far as the apparatus criticus annotation is concerned. The examples in this post are a bit cleaned up, for the sake of clarity.
The place names attested in the chronicle are mainly places where the King’s activities took place, where he passed through or settled or conducted a military confrontation against the Muslims adversaries throughout the nineteen years of his reign. Nevertheless there are also several place names in the chronicle that are not in Ethiopia but which for example mention places with which the Christian kingdom of Ethiopia had a contact as alliance or conflict. There are also many biblical place names, since the chronicler cited abundantly from the Old Testament. All of these have been marked up in the Gǝʿǝz text have been directly linked to place records in the Beta maṣāḥǝft gazetteer.
Whereas the tagging of the place names was much quicker as thought, the major task of this phase was creating a record for those places which were not already in the Beta maṣāḥǝft gazetteer. In this list we had an entry for each of those in the place names analytical index of the Encylcopaedia Aethiopica. These often only contained a title.
To model our encoding of places we have used the good advice of the pelagios community. On one side we have sticked to the Pleiades place concept, we have limited the chronological attestation to periods relevant to the spacial scope, which have been added to PeriodO, and we have encoded information in our TEI files following the Syriaca.org TEI Manual and Schema for Historical Geography. This was a change of major impact on our workflow. And a very positive impact indeed, reflected in our schemas, and in our short day to day usage guidelines on our wiki. Records for places contain the multiple names and location related to that conceptual place record, they can have bibliography and a marked up description. See for example the record for Tǝgrāy which is one of those which were already there, and Bara ʾAǧam, which is one of those added within this project. For those places in the gazetteer for which we have our own coordinates or a link to an entity in wikidata for examples, which does, we also offer a geoJson export (see for example Gondar). The following image shows the record for ʾAdulis.
We have created a record for each place attested in the chronicle which did not yet have a record in Beta maṣāḥǝft and fall within the scope of our gazetteer, i.e. Eritrea and Ethiopia in the most inclusive sense. For place names in the chronicle, which were not part of this domain but are attested in literatures we have used our existing ids and whenever possible we adopted the Pleiades ID. For places which are out of the scope of both our gazetteer and Pleiades we used Wikidata and Geonames ids, with an aim at substituting when possible the geonames with the wikidata item id, which often already contains the first. One example is the reference to the Ottoman empire, for which we use Wikidata Q12560. A second example, the island of Patmos which is attested in the text is marked with a Pleiades ID with its corresponding Gǝʿǝz names, ጥሙስ (Ṭǝmus) as you can see in the example below.
<ab> ወእምዝንቱ፡ ኵሉ፡ ለ<roleName type="title">ማር፡</roleName> <persName ref="http://betamasaheft.aai.uni-hamburg.de/PRS4428Galawdew"> ገላውዴዎስ፡</persName> በዓለ፡ ዝንቱ፡ ዜና፡ ኮነ፡ የዓቅቦ፡ዓቃቢሁ፡ ለኵሉ፡ ከመ፡ <supplied reason="omitted">ተዓቅበት፡</supplied> አሐቲ፡ ብእሲት፡ ወወልዳ፡ እምነ፡ አርዌ፡ ዓቢይ፡ ዘርእዮ፡ ዮሓንስ፡7 በዐለ፡ ራእይ፡ ዘደሴተ፡ በ <placeName ref="https://pleiades.stoa.org/places/599872">ጥሙስ።</placeName> ... </ab>
After all this Mār Galāwdewos, the subject of this story, was protected by the protector of all, just as <was saved> a woman and her child from a great serpent, as John, the subject of the Revelation, revealed it on the Island of Patmos.
The 95% of the task of marking up of places in the chronicles has been accomplished. The mark up also revealed how many times the place names have been mentioned in the chronicle. We have added whenever possible for each of the attestation, dates relative to the chronology of the text. Also in this case we have made explicit a relative reference, by assigning a date in the Gregorian calendar to dates relative to the literary context. The following visualisation, which can only show the places for which coordinates could be gathered clearly shows how these text is fully connected to the entirety of the contemporary world, but it offers also a visualisation in chronological order of the places relevant in each year. Although serving a different purpose the difference from the map above is clear already at this early stage.
This map is actually visualising a KML export feed to the Geobrowser. It is indeed still limited in usefulness, as most of the place references have no location, or just relative location, and those might not be the ones which are dated. But we do not have only that and the geoJson export above. As promised we have developed
- an export of all the places in our gazetteer in the Pelagios interconnection format, on which we would very highly welcome feedback.
- annotations in the Pelagios format for any Textual unit or manuscript catalogue record in our repository.
here is the void http://betamasaheft.aai.uni-hamburg.de/api/placeNames/works which points to two annotations sets here (also paginated) http://betamasaheft.aai.uni-hamburg.de/api/placeNames/works/all and http://betamasaheft.aai.uni-hamburg.de/api/placeNames/manuscripts/all
annotations of individual textual units or manuscripts can be obtained here http://betamasaheft.aai.uni-hamburg.de/api/placeNames/works/LIT3122Galaw where you can see the text of the chronicle of Galawdeos and the same works for manuscripts as well like this http://betamasaheft.aai.uni-hamburg.de/api/placeNames/manuscripts/BAVet1
- there is always a link to the explicit TEI file as well.
Without a reference gazetteer and without annotation of place attestations and names in primary sources it was hard to build a solid ground for the study of historical geography of Eritrea and Ethiopia. We have given this task a first kick by setting up the gazetteer structure and annotating a first text, many problems, probably common to other such efforts for other domains have become evident also for us. The possibilities open, in comparison to historical maps are clear also at this early stage, but there is still a lot of ground research work to be carried out before most of this work can become useful for further research purposes, for the identification of places, the documentation of placename attestations, etc. From the point of view of our current effort we aim at the following for the shorter term.
- Date all placeNames marked up in the text and clear up references inconsistencies.
- We have still plenty to do on the place records which we are actively continuing to work on, adding locations and representative points.
- We want also to mark ethnic names and update the ethnic records we have already.
- We would also like to use some of the data produced with Peripleo and the Pelagios api.
- We will add some visualisations of the data linked to our annotations from Pelagios data to our textual units and places views.
- Also, as we now also have images of manuscripts exposed with IIIF presentation API, it would be nice, as suggested by Rainer, to explore annotations of manuscript images in Recogito.
- It would be nice in the future to have some of the routes most often used on a layer, as well as the extent of the Christian kingdom.