Aksum An African Civilisation of Late Antiquity Stuart Munro-Hay

remains seem rather to show the usual type of `mansion' style of structure for the 
buildings. The church of Abba Pantelewon, outside Aksum, for example, seems to have 
been built on a site already in use in pre-Christian times. There is said to be a stone 
staircase inside it, but this is inaccessible to all but the priests.  
Illustration 56. View of the east end of the old cathedral of Maryam Tseyon at Aksum, 
showing the podium built on the ruins of an earlier, Aksumite, structure.  
The cathedral of Aksum, `Our Lady Mary of Zion', or Maryam Tseyon, was also known 
as Gabaza Aksum, gabaz referring to a church or holy place. Land charters, surviving 
only as later `copies', detail lands purportedly given to the church by Abreha and 
Atsbeha, Gabra Masqal, and Anbessa Wedem (Huntingford 1965). An early account of 
the church's magnificence is said to have been related to the prophet Muhammad during 
his last illness by two of his wives, Umm Habiba and Umm Salama, both of whom had 
been in exile in Ethiopia with their former husbands in the 620s. The description 
mentions that the walls were covered with paintings, and, if so, is testimony to the early 
commencement of a type of decoration which later became standard for Ethiopian 
churches (Sergew Hable Sellassie 186, n. 30). The cathedral was destroyed or damaged 
several times, the last but one before the present structure being a large five-aisled 
edifice, which was eventually destroyed by the Muslim armies under Ahmad Gragn. 
Alvares' description of the church and its environs, extremely valuable as it is the only 
surviving eye-witness description of the ancient church, is as follows (Beckingham and 
Huntingford 1961: 151)  
"a very noble church, the first there was in Ethiopia: it is named Mary of Syon. They say 
that it is so named because its altar stone came from Sion. In this country (as they say) 
they have the custom always to name the churches by the altar stone, because on it is 
written the name of the patron saint. This stone which they have in this church, they say 
that the apostles sent it from Mount Sion. This church is very large; it has five aisles of 
good width and of great length, vaulted above, and all the vaults closed, the ceiling and 
sides all painted. Below, the body of the church is well worked with handsome cut stone; 
it has seven chapels, all with their backs to the east, and their altars well ornamented. It 
has a choir after our fashion, except that it is low, and they reach the vaulted roof with 
their heads; and the choir is also over the vault, and they do not use it. This church has a 
very large circuit, paved with f lagstones like the lids of tombs. This consists of a very 
high wall, and it is not covered over like those of the other churches, but is left open. This 
church has a large enclosure, and it is also surrounded with another larger enclosure, 
like the enclosing wall of a large town or city. Within this enclosure are handsome 
groups of one storey buildings, and all spout out their water by strong figures of lions 
and dogs of stone [of different colours]. Inside this large enclosure there are two 
mansions, one on the right hand and the other on the left, which belong to the two rectors 
of the church; and the other houses are of canons and monks."   
Alvares as one might expect, records that this church, was supposed to have been built by 
queen Candace.  

---

It is presumed that the church of Maryam Tseyon as Alvares saw it reflects the result of 
continual development from an original foundation in or after the fourth century, through 
a number of improvements and additions. A proposed restoration of the church has been 
attempted by Buxton and Matthews (1974), basing their ideas on Alvares' not always 
very clear description. They also embody certain details of architecture from the Lalibela 
churches, which were constructed or decorated according to something of the same 
architectural tradition.  
The Book of Aksum (Conti Rossini 1910) lists a number of churches, some of whose 
names were bestowed, following local tradition, on the palaces which the DAE found in 
1906. The early churches of Ethiopia seem to have been apsidal basilicas, apparently 
following the plan customarily used in Syria (Anfray 1974: 763ff). This contrasts sharply 
to the round churches of the present day, which may well be based on a local African 
architectural tradition, but the province of Tigray is still largely characterised by churches 
of the rectangular plan.  
In Aksum, two basilicas were found on Beta Giyorgis hill, and one was excavated (Ricci 
1976; Ricci and Fattovich 1987). The structure, designated `Bieta Giyorgis Superiore' 
from its position, showed two main building-periods. The lower structure was of typical 
Aksumite construction, but showed a variation in plan as it had side wings flanking the 
north and south walls near the east, or apsidal, end. A later structure was built on the 
original, witho ut the usual Aksumite recessed plan, and consisting of a rectangular room 
whose roof was supported by four stone columns arranged in pairs, and with pilasters 
built in with the main walls on three sides (Ricci and Fattovich 1987: fig. e). In many 
places tombs had been constructed under the paving, and the paving, doorsills and the 
like sometimes included re-used stelae, probably brought from the stele- field found by 
Ricci on top of the Beta Giyorgis hill. In addition, some decorative stonework was found, 
including a block carved in relief with a Greek cross, and various fragments apparently 
from stepped column bases, from a column which was cruciform in section, and from 
water-spouts. A capital was also found, carved with volutes in relief.  
Illustration 57. The relatively modern rectangular church of Maryam Tehot at Edaga 
Hamus stands on the ruins of an Aksumite structure distinguished by the use of cut stones 
in the walling instead of the usual mud-bound rubble.  
A further basilica was excavated at Enda Cherqos (de Contenson 1961). Others are 
known at Matara (Anfray 1974: 756ff), and several at Adulis (Paribeni 1907; Anfray 
1974: 750; Munro-Hay 1989i). Further such churches have been found at Agula and 
Qwiha, and very likely the surviving structures at Tekondo and Qohayto include churches 
(Doresse 1957: 200-201). An early (sixth-century? — Doresse 1957: 231-2) church was 
also constructed inside the temple at Yeha. Doresse speculated as to whether the internal 
circular colonnade of one of the Adulis churches excavated by Paribeni (1907: fig. 50) 
might represent an early example of a tendency towards the circular plan. Paribeni 
himself thought that the circular pillared structure in the church was a later addition, 
perhaps the support for a wooden baldaquin over some now-removed object. It may be 
that during the centuries there were some changes in the church ritual which were felt to 

---