Michael Gervers


The Zaguë capital, Lalibäla, with its twelve rock-cut and monolithic sites designated as churches, has received more attention in travel and scholarly literature than any other medieval place in Ethiopia. The churches known as Däbrä Sina and Golgotha (Fig. A) have been considered of primary importance within the rock-cut complex since it was first described in western literature by the Portuguese priest, Father Francisco Alvares, following his visit in 1521.[1] He describes ten churches[2] in considerable detail, and ends by saying “I weary of writing more about these buildings, because it seems to me that I shall not be believed if I write more ….”[3] More than four hundred years later, A.A. Monti della Corte referred to the site as “la gemma archeologica e artistica indiscussa”.[4]

A third important component is the small, totally rock-cut structure described almost universally in the literature as the Sellassie Chapel or “Crypt”, which is accessed through a simple doorway at the east end of Golgotha’s south aisle (Fig. A:H). It is, in fact, no more a crypt than any other rock-cut monument in Lalibäla because it is on the same underground level as the Däbrä Sina and Golgotha churches, and has no vertical link to the surface. It is, however, an exceptionally dark spot, since natural light comes only through a single, small window in the SW corner, and is held to be such a holy place that few people outside the superior ranks of the priesthood have ever been granted access to it. The first non-ecclesiastics known to have entered this chapel were Italians in 1939, who shortly afterwards provided detailed descriptions.[5] According to the Memhir with whom Beatrice Playne spoke in 1946 or 1948, this sacrilege took place “at the point of a pistol”.[6] Since that time, access has occasionally been afforded to laymen, [7] who have described the Däbrä Sina– Golgotha – Sellassie complex as “Lalibala’s most secret and holiest place”[8] and the Sellassie chapel as “the place of greatest sanctity in Lalibela”.[9] Bianchi Barriviera considered the tri-partite unit as unique “and perhaps the most important for the study of monolithic churches”.[10]



The unique nature of the rock-cut monuments in Lalibäla, together with their sanctity and, especially in the case of the Sellassie Chapel, its secrecy, has led some scholars to go well beyond the evidence in their attempts to interpret the “meaning” of the site as a whole.  Two particularly imaginative examples come to mind, the first being the widely distributed work of Irmgard Bidder. In Lalibela, The Monolithic Churches of Ethiopia, published in 1958, she theorized that the structures as we know them today are reworked versions of ancient sanctuaries originally devoted to nature worship. This line of reasoning led her to many quite fantastic conclusions, not the least of which was her overall view of the cosmology of the site. It was, she thought, the physical expression of a creation myth. She considered the churches in three groups, the first concentrated in the NE section, near the monolithic church of Betä Emanuel. They, she wrote, represented the “Womb of the Earth”. The second group, concentrated around the Grave or the Church of Golgotha in the NW section, represented “the form and the idea of the stele projected horizontally into the rock’s surface”; in other words the impregnating phallus. The remaining unit, the single church of Betä Giyorgis to the SE, expressed “logically the idea of the ‘Offspring of Heaven and Earth’”.[11] In the nearly half century since the book was published, perhaps because it was so fanciful, no one has commented upon her reasoning, although the work is frequently consulted for its ground plans (“adapted from Monti della Corte”) and cited for its descriptions and illustrations.

Another comprehensive hypothesis was proposed by the much regretted Jacqueline Pirenne in her paper “La signification symbolique des églises de Lalibéla, à partir des inscriptions découvertes en 1980-1983”, delivered at the Eighth International Conference of Ethiopian Studies in 1984.[12] Inspired by the discovery and publication by Gigar Tesfaye of four inscriptions engraved on fragments of wood and bearing the theme of Christ’s Transfiguration on Mount Tabor, and of the invocations inscribed on ten mänäbert tabot of which nine were attributed to King Lalibäla,[13] she concluded that the excavated site, for which she gave credit to the king alone, represented a mystical commentary on the Apocalypse of St. John.[14] Much of her argument derived from the association of each tabot with a specific church (identified by similarities in the decorative elements common to both the tabot and church), and from her absolute conviction that the attributions to King Lalibäla were original and correct. However, Stuart Munro-Hay has recently expressed serious doubts about the matter, suggesting that the mänäbert tabot may be no more than “pious forgeries dating from a subsequent phase”, and pointing out that even if they are the Zaguë monarch’s work, “they do no more than indicate that King Lalibela took an interest in the churches – they are far from proof that he had all, or any of them, constructed.”[15]

            The idea that King Lalibäla might not have been responsible for all, or indeed any, of the rock-cut and monolithic structures at this eponymous place, has been growing steadily over the years.[16]  It is not necessary to return to Bidder’s theories to find precedents for rock-cut religious architecture in the highlands of Ethiopia. In 1972, Taddesse Tamrat proposed that the Zaguë king, Yəmrehänna Krəstos, was responsible for introducing the phenomenon in the mid-twelfth century,[17] while Claude Lepage  suggested simultaneously that the rupestrian “églises de vallée” in Tigray, such as Degum, were totally Axumite in their rendition, that they very clearly preceded the examples from Lasta, and that some of them might well have been executed as early as the seventh century.[18]  There is no absolute proof either way, but there can be no question that Axumite elements appear to a greater or lesser degree in the rock-cut churches of Ethiopia, and that proportionately more of them can be found in the churches of Tigray, the traditional centre of Axumite power and influence, than in Lasta and regions to the south.



Developing a chronology for Ethiopia’s rock-cut ecclesiastical heritage is essential and scholars like Lepage, and David Buxton before him,[19] have paved the way through their sometimes meticulous analysis of the physical make-up of these monuments and of the perceptible similarities and differences among them. One must exercise caution in attempting to determine the age of a rock-cut or monolithic structure, however, because unlike a built counterpart, one cannot trace its history back any further than its visible surfaces.  Any earlier stage of construction is by necessity removed forever by the workman’s chisel. As a consequence, it is impossible to determine whether the rock-cut monuments of Lalibäla ever existed in a form other than what can be seen at present. The question, as far as the churches of Lalibäla are concerned, is whether they were all constructed under the aegis of King Lalibäla in the twenty-two or twenty-four year period Alvares understood it took to make them.[20] Certainly, the job could have been done in that time, for the tufa from which the monuments were cut would not have been particularly difficult to excavate before it was exposed to the air;[21] yet the structures display a sufficiently wide range of architectural and stylistic differences to make it doubtful that they all belong to such a short period of time. Buxton has suggested that Lalibäla represents “a museum of church types”,[22] but while that may actually be the case it can hardly have been the original objective of its constructors. One may also wonder for what purpose they carved out the complex in the NE section of the site. In comparison with the structures in the NW section, only Betä Emmanuel and Betä Libanos are oriented; the others may once have served for civil purposes and should probably not be considered in terms of ecclesiastical architecture at all.[23] We remember that the fifteenth-century gadle of King Lalibäla reports that the angel of the Lord instructed the king to build ten churches.[24] Eleven dedications are actually cited, but Betä Lehem is not included among them, quite probably because it was not a church at the time. There is, furthermore, the possibility, alluded to by several scholars,[25] that some at least of the dedications have changed over time.



If there is any correlation between the presence of Axumite elements in Ethiopian rock-cut architecture and the relative antiquity of any given site, then Betä Maryam (Figs. B/E:O), Abba Libanos, and Betä Emmanuel, and perhaps in its enlarged formula, Medhane Alem, represent the most conventional type among the churches of Lalibäla. Excluding Betä Merqorios, Betä Gabriel and Betä Lehem, which are not oriented and which may as a consequence never have been conceived as churches,[26] the least “Axumite” of the ecclesiastical structures in Lalibäla are those of the Däbrä-Sina - Golgotha – Sellassie complex, and Betä Giyorgis. They in fact appear to disregard Axumite traditions almost entirely, notwithstanding the non-functional transversals in the entrance door and lower range of windows at Betä Giyorgis (Fig. 1).[27] The absence of long-established forms suggests that these monuments were carved at a time when those traditions were no longer considered essential in church construction; that is, at a date following the creation of Betä Maryam and others of its type. In terms of the ground plan alone, that so-called Axumite type consisted of a central square divided into three aisles by two sets of two pillars contained in a rectangular frame (Fig. B:O).[28] At each end of the central square this frame incorporated a tripartite division of space corresponding to the layout of the three aisles. At the east end, the sanctuary stood as a continuation of the central aisle, and was flanked to the north and south by rooms, accessed from the sanctuary and/or the side aisles, which served as sacristies.  At the west end, a central entrance vestibule was situated between two chambers, with access to the northern chamber from the vestibule and to the southern from the south aisle. In addition to the west entrance, there were northern and southern ones which opened into the westernmost bay of the central square. This is precisely the formula found at Betä Emmanuel and Betä Libanos,[29] and very close to that of Betä Maryam which is otherwise distinguished by two additional pillars, one in front of the sanctuary and the other opposite in the entrance vestibule.



The only means of entering the Däbrä Sina – Golgotha – Sellassie complex is through the south or west portals of Däbrä Sina (Fig. A). The interior measurements of Däbrä Sina are approximately 9.50m E-W and 8.50m N-S. The church is divided into three aisles by two rows of four cruciform columns which support arches rising to a height at mid-point of about 3.4m. The flat ceiling is 5m high.[30] Unlike the examples described above, there are no chambers in the four corners of Däbrä Sina. The pillars at the east and west corners of the central square are free-standing, not engaged, thereby doubling the usual number from four to eight. In another divergence from Axumite principles, the east end, which is wholly occupied by the sanctuary, is accessed from the north and south aisles by a step rising from the back of the first set of pillars, and from the nave by a double step rising between the second set of pillars (Figs. A/C:A).[31] As a result, the clergy have a considerably larger space for their use during the celebration of holy offices.



Narrow doorways in the NE and NW section of Däbrä Sina’s northern wall, leading out of the first and fourth bays, serve as the only entrances to what is known as the Church of Golgotha (Figs. A/C/D:B & C, 2). Compared to other churches in Lalibäla, and noticeably to Däbrä Sina, this is a poor structure, badly wrought, yet considered to be among the holiest places on the site because it houses the so-called “Tomb of Christ” and, supposedly, the remains of King Lalibäla himself (Figs. A/C/D: F,G).  Its 10.7m E-W length is separated into two aisles by three cruciform piers. The total width measures approximately 6m. The north aisle is wider than its southern counterpart, suggesting that the original intention may have been to create a tri-partite structure with a nave broader than each of two flanking aisles. Certainly, compared to the southern wall, the northern wall is rough and unfinished, leaving the impression that it was hastily prepared. An entablature in low relief carries around the south, east and west walls at the point where, on the pilasters, arches spring from bracket capitals to corresponding brackets on the three central piers (Fig. D). Arches also link the piers to each other on the E-W axis of the structure. Unlike those in Däbrä Sina, the capitals are devoid of any chiseled decoration. Painted bands, in greenish-blue and red, and probably dating from the 19th-century, run along the south wall at the level of brackets and capitals. A greenish-blue rectangle surrounded by a triple frame of bands in the order of yellow, red and yellow, covers the north face of the pilaster capital standing to the east of the door connecting Golgotha to Däbrä Sina in the third bay. The bracket above it is painted in an unsuccessful attempt to create a Greek meander. The pier below bears a representation of a female saint in frontal position. The western wall, reached by a step which in the north aisle is broad enough to serve as a plinth (Fig. A:E), is pierced by windows on two levels (Figs. D, E). At the upper level in each aisle the opening is semi-circular and corresponds to the transversal arches which spring from the north and south walls to meet at the central row of piers.  The entablature separates these openings from five cruciform windows which pierce the wall below, two in the south aisle and three in the broader north aisle. Each of the five windows is itself set back into the top of a vertical recess which stands between the raised floor and the entablature.

At the opposite end of the interior, in the NE corner, the first bay of the north aisle rises three steps above the floor of the church. A blind archway in the east wall of this bay may have been intended to contain a monumental standing figure in relief, but nothing can be discerned from available photographs.[32] A final transversal arch rises above it from within the wall creating a blind lunette below.

A recess in the north wall, protected by an iron grate, is designated the “Tomb of Christ”. Within, a recumbent figure in low relief lies at floor level, facing east (Figs. A/C/D:G, 3, 4). The body tapers from broad shoulders to his feet, which are integrated into the east wall of the arcosolium. His hands are crossed over his chest and he wears a short, thigh-length tunic. The face is without features and was either unfinished or rendered purposely thus. An angel stands in relief at his feet; if there was once one also at his head, as postulated by Barriviera,[33] it was not visible to this writer.[34]

The second bay of the north aisle contains the “Tomb of Lalibäla”,[35] surmounted by a large mänbärä tabot draped in cloth. Today, it, together with the raised first bay, is normally curtained off.[36]

In the spaces between the pilasters along the north and south walls of Golgotha are five larger-than-life-sized figures standing in relief within blind arches (Figs. A/C/D:2-6,  5).[37]  Those in the north wall appear, like the wall itself, to be unfinished (Fig. 6). Each is identified as a saint by inscriptions on the front of the arches, but since these inscriptions are only lightly incised and the letters unevenly distributed along the curve of the arch, it is generally thought that they were added at a later date.[38] Similar arches over the narrow doorways leading from the south aisle into the NE corner of the Church of Däbrä Sina on the one hand and into the Jesus Cell on the other, suggested to Barriviera that two more saintly figures in relief may once have stood below them, but that these were lost when the doorways were cut through.[39]



Two doorways open from the SE corner or first bay of the south aisle, one leading through the south wall into a small chamber measuring 2 x 1.3m known as the Jesus cell (Fig. A:D) and the other through the east wall to the Sellassie Chapel. The Jesus cell is linked to the east end of the north aisle of Däbrä Sina by a small aperture in its SW corner. A modest window with a pointed arch looks out from high in the wall of the SE corner over the deep rock-cut passage which delimits the east end of Däbrä Sina. An interior frieze is said to be decorated with leaves similar to those which extend like swans’ necks out of the upper frame of the broken arch on the exterior wall (Figs. A:J, 7).[40] According to Barriviera, this cell formerly contained a tabot.[41]



The Sellassie chapel is a rough trapezoid measuring 6.5m wide at the east end and 4.6m at the west. It is 6.8m deep on the north side and 6m on the south. The south side was made shorter to avoid cutting into the pre-existing Jesus cell (Fig. A). A single, 60cm-square, 5.4m-high pillar rises on the central axis about a third of the way into the room. It has no capital, but brackets support wide ribs which cross transversely and longitudinally above it. A 40cm-high step, or plinth, leads eastwards out of the base of the pillar for a distance of 1.5m, whereupon it intersects with, and continues a step higher than, three well-delineated steps leading up to a platform which extends on a slight convex curve from one side of the room to the other. In the centre of the platform and directly in line with the column at a depth of 50cm from the top step is a monolithic altar, flanked by two others equi-distant from the north and south walls (Figs. A, C, 8). The central monolith is 1.5m high, while those on either side stand at 1.35m. Each is roughly 70 to 90cm square. The vertical surfaces are divided horizontally with crosses incised at the upper level on each of the four faces. Anthropomorphic symbols of the four beasts of the Apocalypse, with arms raised in prayer, fill the lower faces of the central monolith: the man in the west, the eagle in the east, the bull in the south and the lion in the north.[42] Corresponding surfaces in the flanking monoliths are empty. In the east wall behind the central monolith is a 2.7m high niche surmounted by an arch.[43] The niche, 1.4m wide, is stepped on three levels at the bottom, increasingly recessed to a depth of 1.1m before the springing of the arch.  In a blind arch on the north side of this central niche, directly behind its corresponding monolithic altar, a larger-than-life anthropomorphic figure with the head of an ox stands in relief (Figs. 9, 10). The hands are held clasped above the waist, while the head is encircled with a nimbus containing a cross in the top centre. Monte della Corte’s description of this head as that of an ox, although true, has been much derided by subsequent commentators.[44] The decoration on the drapery of this figure seems somewhat more elaborate than on the well-preserved figures along the south wall of Golgotha, but all would appear to be contemporary. The figure intended for the south side is represented only by a small fold of drapery in the lower left corner of the space delineated by the arch, apparently all that the sculptor achieved before his work was definitively interrupted.

At the bottom of the steps on the north wall is the opening to an empty tomb chamber. The west end of the south wall is pierced by a single small window which, from the outside, appears next to, and is framed with decoration very similar to, that of the window opening into the Jesus cell (Figs. A:I, 7).[45] The swan-necked, floral pattern which delineates the frame of these windows with pointed arches is again reproduced in all twelve of the upper range of windows in the Church of Betä Giyorgis, although in the latter they are not so delicately conceived (Fig. 1).



The Däbrä Sina-Golgotha-Sellassie complex is certainly unique among the Lalibäla churches in its structure and composition, devoid as it is of nearly every traditional characteristic of Axumite-style architecture. The absence of corner chambers in Däbrä Sina is a new departure in church construction, with eight piers instead of four (Fig. A), thus creating an open interior space considerably larger than that of an Axumite precursor constructed around a square. There is no evidence that there was any intention to excavate the rest of the complex when the Church of Däbrä Sina was conceived. On the contrary, the narrowness of the two doorways in the north wall of Däbrä Sina which are the only means of entry to Golgotha, suggests they were cut at a time when the architectural vocabulary of the former had already been established. Golgotha itself, with three free-standing piers, only two aisles and also devoid of corner chambers, is even less conventional in its conception than Däbrä Sina. Its construction preceded the next stage of expansion which was the tiny Jesus cell.[46]

The final excavation, and perhaps the last of any of the rock-cut monuments in Lalibäla, was the Sellassie Chapel. It must have followed the completion of Golgotha because it can only be entered through the doorway at the east end of Golgotha’s south aisle; and it must have followed the excavation of the Jesus cell as its SW corner is excavated on an angle in order not to cut into and obliterate the cell’s east wall. Thus, there is a very clear progression in the construction of the complex. As Däbrä Sina is the oldest of the group and already devoid of Axumite features, it is not surprising that such features are also totally absent from the rest. There are no transversals (square “monkey heads”), typical of Axumite architecture, to be found in any of the doors and windows. The windows from the exterior are barely framed, if at all, and the relief of those which are decorated is shallow (Fig. E). There is, furthermore, no consistency between the shapes of the windows in Däbrä Sina and those in Golgotha, not to mention the two with highly stylized meandering fronds which distinguish the Jesus cell and Sellassie Chapel from the others (Fig. 7). They, too, could have been cut as an afterthought. The amount of light they let through is minimal and their placement is entirely utilitarian. These windows can, therefore, be considered to be among the newest of the new.



Not to be overlooked when considering the windows of the complex is the Axumite doorway opening onto the top of the deep trench delineating the west exterior wall of the Church of Golgotha (Figs. B/E/F:Q, 11). This doorway, described in the literature as a “balcone[47] and understood to be a window, stands at the western extremity of a corridor which is directly in line with the E-W axis of the Church of Betä Maryam, and which opens through a double portal into the courtyard before the west façade of that church (Figs. B/C/E/F:M). The corridor is now open to the elements, but it is very likely that it was once a rock-cut passage and that the roof collapsed. It is also probable that this corridor served as the principal access to Betä Maryam before the deep trench surrounding the Däbrä Sina-Golgotha-Sellassie complex was cut.[48] That would have been even before the carving out of the so-called “Tomb of Adam” onto which the Axumite doorway looks (Figs. A/B/E/F:K, 11). It is, in fact, almost certain that this trench was once far shallower and that formerly it was the principal link between the river now known as the Jordan and the entrance to Betä Maryam.  At that time the site was still designated as Roha, presumably derived from al-Ruha, the Arabic name for the ancient riverine town of Edessa in Asia Minor. Edessa is a Macedonian term referring to “abundant water”. [49] It has recently been argued that the etymology of al-Ruha can be traced to the Arabic rawahah, meaning “low plain where water accumulates”.[50] It is this association which may lie behind the choice of the name Roha for a site with plentiful supplies of water. The now functionless doorway opening onto the void may thus be seen as a solitary witness to the Betä Maryam site before the excavation of the present deep trenches leading both to the river and around what subsequently became the Däbrä Sina-Golgotha-Sellassie complex.



If the concept and aspects of the architectural and decorative details of the Däbrä Sina-Golgotha-Sellassie complex are unique to Lalibäla, they are not unique to Ethiopia. In his consideration of the chronological development of ecclesiastical architecture, and especially of rock-cut ecclesiastical architecture in Ethiopia, Claude Lepage noted already thirty years ago that there was a clear progression from late Axumite (or early Zaguë),[51] to various stages of Axumite-influenced, and then to post-Axumite styles. Among the first, he refers to what he calls the “églises de vallée” of which an early example is Degum;[52] the second, to Berakit, Čerqos Wuqro, Abreha-Asbeha and Mikael Amba,[53] which relate to the churches built in caves by the Zaguë and which (with the exception of the possibly contemporary Berakit) precede the late Zaguë monoliths in Lalibäla;[54] and third, to the post-Zaguë period which saw the construction of such rock-cut churches as Enda Maryam Wuqro (Amba Sanayt) and May Kado Giyorgis  in Tigray in the late-thirteenth to early fourteenth century and thereafter.[55] The latter are of particular significance in the present study because both have a raised, central sanctuary containing a monolithic mänbärä tabot and a raised platform in the NE corner. In the case of May Kado, this platform contains a monolithic altar, while at Enda Maryam a similar altar is situated in a room of modest dimensions entered from the north side of the central sanctuary.[56] The west faces of both bear a cross in relief closely resembling those in the upper sections of the three monolithic altars in the Sellassie Chapel (Figs. 8, 12). The raised position of the NE corner of these two churches further recalls that of the Church of Golgotha (Fig. C), and points to the probable use of the area in all three cases as an extension of the sanctuary. The monolithic altar in May Kado confirms this employ. It has already been proposed that the Sellassie Chapel in Lalibäla, although constructed at a somewhat later date, served as a sanctuary for the Church of Golgotha. The central sanctuaries with monolithic mänäbert tabot may also have been later additions to the other two churches. There is, furthermore, a small chamber leading off the north side of the NE corner in Maryam Wuqro which may have served the same purpose as the arcosolium containing the “body” of the Dead Christ in Golgotha.[57] Such similarities do not presuppose a direct influence between these arguably post-Axumite style churches in Tigray and the Däbrä Sina-Golgotha-Sellassie complex at Lalibäla in Lasta, but they do have enough in common to suggest that they are related chronologically. Fundamental to this argument is the presence of the raised sanctuary, a feature common to all of the aforementioned structures, as well as to Lalibäla Betä Giyorgis. Lepage’s proposed late thirteenth- or early fourteenth-century date for the Tigrayan monuments is difficult to confirm, but the reduction, or total absence, of Axumite architectural elements and traditions in all of these churches does point to a post-Axumite period for their excavation in the rock.[58]



The presence of monolithic altars is of prime importance to the argument for the particularly late construction of the Sellassie Chapel. Their appearance at Maryam Wuqro and May Kado Giyorgis in the context of the raised sanctuary has already been mentioned, but there are many more in Tigray and the more one looks the more one finds. Of fourteen examples cited here, it is certain that ten churches have a monolithic altar or mänbärä tabot in the central part of the sanctuary.[59] If there is a second altar, and there often is, it is usually to be found in the NE corner of the church,[60] confirming Lepage’s assessment of the ceremonial importance and sanctity of this part of the monument.[61] Abba Yohanni has a monolithic altar at the east end of each of its three aisles (Fig. 13). Altars of this type in the SE corner of Iyyesus Wanza and Mikael Barka suggest that each may also have a total of three. Iyyesus Wanza contains interior features remarkably close to those at Däbrä Seyon, which also contains a monolithic altar and is dated to the period 1382-1411.[62] In chronological terms, churches with only one monolithic altar, as at Maryam Dengelet, could represent the earliest of their type. These may have been followed by those with similar monoliths at the east end of the north or south aisles when those aisles terminate to the west of the sanctuary, as at Iyyesus Wanza and May Kado Giyorgis (Fig. 12). Those with monolithic altars on the same plane to the north and/or south of the central sanctuary, as at Enda Maryam Wuqro, Mikael Barka and Tselal Moo, may either be later extensions to a church which began with a single monolith in a central sanctuary “apse”, or were carved into the rock at a time when the popularity of such altars coincided with the enlargement of the sanctuary area to the full width of the church, as was probably the case for Abba Yohanni. At Abräha Atsbäha, an altar constructed of stone and mortar in the NE corner of such an expanded sanctuary, may represent a later attempt to keep up with the new custom.[63] Tall monoliths, measuring up to 2.7m in height and fashioned to serve as the mänbärä tabot in the central sanctuary area of some churches may well belong to the latest period of development, and in some cases have been introduced as an eastward extension to the sanctuary of a previously existing rock-cut church (Fig. 14).[64]

            All those who have attributed dates to churches with a monolithic altar or mänbärä tabot place them between the late thirteenth century (starting with the reign of Yekuno Amlak, 1270-1285) and the early fifteenth century;[65] that is, to the post-Zaguë period. Significantly, the one church known to have three such monoliths is Abba Yohanni, a late structure believed to bear the name of the fourteenth-century Abba Yohanni whose Vita was composed in the fifteenth century by Menas IV, bishop of Axum.[66] Seen in this context, the three carefully worked monolithic altars in the Sellassie Chapel would appear to be the high point of this rupestrian development which, probably in the late fourteenth to early fifteenth century, filtered its way down from Tigray to Lalibäla in Lasta.



An anomaly in any style and at any time is the presence in the wall niches of the Church of Golgotha and the Sellassie Chapel of larger than life-sized relief sculptures representing holy figures (Figs. A/C/D:1-6, 5, 6), and of the recumbent figure in the arcosolium in the NE corner of Golgotha said to represent the Dead Christ (Figs. A/C/D:G, 3, 4). All those who have studied them comment that they are unique not only in Lalibäla, but in Ethiopia as a whole. While the identity of the standing figures remains uncertain, there can be no doubt that they represent saints since all but one has a nimbus. That exception is the most frequently reproduced because not only is the relief in nearly perfect condition, standing below the small window opening into the Church of Däbrä Sina in the centre of the south wall of Golgotha, but the figure is depicted wearing a turban (Fig. 15).[67] The inscription on the arch above identifies him as St. Cyriacus, but since, as was mentioned before, the inscriptions do seem to have been added later, this information may probably be disregarded. Furthermore, the presence of a turban rather than a nimbus suggests that the figure is not meant to represent a saint, but instead a high ecclesiastical official. It is also notable that this relief is small in comparison to the others, perhaps to indicate a figure of lesser stature than a saint. Such differences do detract from the theories that these reliefs represent the Nine Saints or the Twelve Apostles.[68] Elsewhere, from a chronological viewpoint, the introduction of oriental dress goes hand in hand with the rise of the Ottoman Turks and the spread of their empire in the fifteenth and sixteenth centuries. The Ethiopian Court had adopted Turkish court costume with narrow sleeves by the sixteenth century and in 1628 Almeida recorded that “Rich men have [round, red] caps worked for them in silk and gold by Turkish tailors”.[69] The turbaned representations of the saints and biblical figures in the church of Abba Yemata at Guh (Gäralta, Tigray), perhaps the closest in style to the Golgotha figure, have been attributed broadly to the fifteenth to seventeenth century (Fig. 16).[70]

All scholars note that the reliefs are unusual. Sergew Hable Sellassie points out that “the usage of statues in the church is not at all a practice of the Ethiopian church nor of the Orient in general. It is reminiscent of the western practice.”[71] Similarly, Jules Leroy finds the presence of these “sculptures animés” to be “un fait très rare et contraire à la pratique générale des chrétientés orientales”,[72] and comments with reference to the recumbent figure that in a western context the representation of the Entombment in churches does not become widespread until the fifteenth century.[73] Lepage compares their style to examples in “divers manuscrits bien datés” of c. 1400.[74]  The general consensus points to a fifteenth century date for these reliefs.[75] A postquam date is provided by Alvares who reported seeing two of them in 1521.[76]



Despite the presence of the relief sculptures, the Church of Golgotha is hardly a burial place fit for a thirteenth-century Zaguë king, not comparable in quality with such other churches in Lalibäla, as for example Betä Maryam, or even Däbrä Sina. Had King Lalibäla chosen his own burial place, one might have thought that he would have provided something better for himself. But if the evidence here is correctly interpreted, there is little possibility that the Church of Golgotha even existed at the time of the king’s death, and obviously even less that he was buried next to the symbolic tomb of Christ when that tomb with its recumbent effigy may not have been fashioned from the rock for up to two centuries after his death. Either his remains have been translated from elsewhere,[77] or Lalibäla may not be buried in what for at least the past five hundred years has been referred to as Lalibäla’s tomb.[78] Furthermore, it has been recently argued that it was only from the fourteenth century that the churches at Lalibäla began to be ascribed to the king in Ethiopian sources,[79] and not until the fifteenth century that “pilgrimage to the tomb of Lalibela developed”.[80] The fifteenth century is also the time when images of King Lalibäla first appear in art.[81]

This chronology suggests that for reasons still unclear, the kings of the Zaguë dynasty, long branded by scholars as usurpers, were rehabilitated in the fifteenth century. This is the century when, as far as can be determined, the names of the Zaguë kings Lalibäla, Neakkuto Leab, Yitbarek and Harbay (i.e. Gebre Maryam) were first included in the Ethiopian version of the Synaxarium, or Book of Saints.[82] Their inclusion is a clear indication that they and their dynasty were then acceptable to the Orthodox Church. This is also the century when the lives of Lalibäla and Neakkuto Leab entered the literature as hagiographical texts,[83] and the century when the name Lalibäla is first used to designate the site of the rock-cut and monolithic churches which the Zaguës had called Adefa or Roha.[84] The evidence for the incorporation of the kings of the Zaguë dynasty, and especially of King Lalibäla, into the mainstream of Ethiopian religious culture in the fifteenth century is overwhelming. The process may have begun in the previous century when, for example, the authors of the Kebra Nägäst gave Solomonic legitimacy to the Zaguë dynasty by tracing their descent to the handmaiden of the Queen of Sheba who, like the Queen herself, was impregnated by King Solomon.[85]

One may ask why this change which was so supportive of the Zaguë came about. There is no firm answer, but it could be suggested that it was related to the long-standing controversy over the celebration of the double Sabbath which had been so disruptive throughout the land during most of the fourteenth century, and even before, only to be resolved to the satisfaction of the followers of the monk Ewostatewos by King Zar’a Ya’qob at the Council of Däbrä Metmaq in 1445.[86] The reasoning behind this association lies in the text inscribed on a mänbära tabot, now in the Church of Däbrä Sina, attributed to King Lalibäla and designated by Gigar Tesfaye as “Golgotha 1”.[87] This perfectly preserved, lengthy inscription is consecrated to Sunday, “the sabbath of the Christians”.[88]  Such a statement of orthodoxy may possibly belong more to the Alexandrian position of the fourteenth and fifteenth centuries than to a Zaguë king of the early thirteenth, unless of course, the observance of Saturday was already rooted in the Zaguë period.[89] This mänbära tabot, together with four others from the Däbrä Sina-Golgotha-Sellassie complex, two from the Church of Mädhane Aläm and two from Betä Gabriel, are thought to have been produced in the early thirteenth century, but although they all bear the name of King Lalibäla, there is nothing to prevent their having been made at a later date.[90]

One may suggest, therefore, that the kings of the Zaguë Dynasty were rehabilitated by the Orthodox Church in the fourteenth and fifteenth centuries possibly in an attempt to counter the Tigrean heresy of Ewostathianism which at that very time was penetrating Lasta from the north. The Zaguë capital at Roha had by then been abandoned as a centre of government, but the Axumite-style monoliths and rock-cut churches remained and, like the foundation of St. James of Compostella in Spain shortly after the Muslim conquest of the Iberian Peninsula in the eighth century, circumstances encouraged their transformation into a centre of orthodox pilgrimage. A pilgrimage site needs a saint, however and, possibly because he had some reputation as a supporter of orthodoxy, King Lalibäla was selected for this purpose. His relics were retrieved, or invented, and an especially dark, mystical space was hastily carved out of the rock to house them. Lasta was former Zaguë territory and in the fourteenth century both the Church and the monarchy may have sought to engender local support by promoting the memory of, and sanctifying, a popular Zaguë king who may himself have triumphed over an earlier wave of “Sabbathic” monasticism.

One need not insist upon this interpretation of the motivation which lay behind the sanctification of King Lalibäla and the creation of the Däbrä Sina-Golgotha-Sellassie complex, but it cannot be denied that the name of the king was introduced to the Synaxarium at about this time, and that the rock-cut complex incorporates architectural features and liturgical elements which postdate those apparent in the other ecclesiastical monuments on the site. The period witnessed a vogue in the production of mänäbert tabot. These may include those made from wood and bearing invocations ascribed to the king, those in the very same style carved as monoliths in the Sellassie Chapel (Fig. 8) and, the largest of them all, the so-called “Tomb of Adam” (Fig. 11) and the Church of Betä Giyorgis (Fig. 1) which, as David Buxton so aptly put it over forty years ago, “is nothing more nor less than a glorified portable altar (manbar)”.[91] The characteristics, both decorative and architectural, which Betä Giyorgis holds in common with the churches, chapels and cells of the Däbrä Sina-Golgotha-Sellassie complex, make it quite clear that it is contemporary with them. The “Tomb of Adam”, in turn, can be compared with the probably late medieval Tigrayan phenomenon of the monolithic mänbärä tabot, while the positioning of the cross on the upper level of the south face recalls those on each of the monolithic altars in the Sellassie Chapel. In this case, however, the shape of the cross suggests an even later date, further confirmed by the fact that the lower half of this monolith extends below the floor level of the Golgotha complex, and was almost certainly fashioned out of the rock after its completion. As for the life-size relief sculptures in the Church of Golgotha, perhaps the figure wearing the bulbous turban (Fig. 15) is the abuna who sponsored the construction work, or maybe even the Patriarch of Alexandria who represented the orthodoxy which this holy sanctuary was created to promote.




The author is grateful to Professor Alessandro Bausi of the Istituto Universitario Orientale in Naples for his valuable reflections on the content of this paper; to Ato Kebede Amare, Commissioner of Tourism in the Tigray, and to Professor Stanislaus Chojnacki for helping with the identification of churches containing monolithic altars; to Dr. Nicoletta Barbarito of the Canadian Embassy in Rome and to Dr. Livia Varga for assistance in the translation of various texts; to Paul Henze for permission to reproduce his photographs as figs. 3 & 4; to Ken Jones at the University of Toronto at Scarborough for preparing the illustrations for publication; and to Gillian Long of the DEEDS Project, University of Toronto, for bibliographical research and editing.


[1] C.F. Beckingham and G.W.B. Huntingford, eds., The Prester John of the Indies. A true relation of the lands of the Prester John, being the narrative of the Portuguese Embassy to Ethiopia in 1520 written by Father Francisco Alvares, 2 vols., Hakluyt Society, Works, ser. 2, nos. 114-15, Cambridge, 1961 (hereafter: Beckingham & Huntingford, Prester John), vol. 1, pp. 205-28, esp. pp. 205-21.

[2] The gadl or Vita of King Lalibäla speaks of an angel instructing the king to build ten churches (J. Perruchon, trans., Vie de Lalibala Roi d’Éthiopie, Publications de l’École des Lettres d’Alger, Bulletin de Correspondance Africaine, Paris, 1892, pp. 121-27 (hereafter: Perruchon, Vie de Lalibala); also quoted in Gabriel Simon, Voyage en Abyssinie et chez les Gallas-Raias. L’Éthiopie, ses mœurs, ses traditions, le négouss Iohannès, les églises monolithes de Lalibéla, Paris, 1885, pp. 321-22). There are several differences in these two accounts: only the Vie de Lalibala includes the churches of Däbrä Sina, Betä Gabriel and Betä Libanos, while only Alvares mentions Bethlehem and “Lalibela”, which latter he cites as “the principal one”. Both include Golgotha, by which Alvares may also have meant Däbrä Sina (see below). Neither mentions Mikael, which appears to be a later appellation for either Golgotha or Däbrä Sina (In L. Bianchi Barriviera, Le chiese in roccia di Lalibelà et di altri luoghi del Lasta, extract from Rassegna di Studi Etiopici, vols. XVIII-XIX (1962 & 1963) (hereafter: Barriviera, chiese in roccia) , pl. 8 & 8bis, Däbrä Sina is described also as Chidane Meret, while the western part of what most refer to as Golgotha is described as Mikael and the eastern part as Lalibäla. His plan places Golgotha in the western part of what has become known as the Sellassie Crypt, and Sellassie in the eastern part. On the other hand, Alessandro Augusto Monti della Corte (Lalibelà. Le chiese ipogee e monolitiche e gli altri monumenti medievali del Lasta, Rome, 1940 (hereafter: Monti della Corte, Lalibelà) , p. 56) and Irmgard Bidder (Lalibela. The monolithic churches of Ethiopia, Cologne, 1958 [hereafter: Bidder, Lalibela] , fig. 25, p. 117) identify Däbrä Sina as Mikael and make no reference to Lalibäla or Chidane Meret). Neither the Vita nor Alvares makes specific mention of the Sellassie Crypt and it may well be that Alvares never entered it.  The absence of any reference in the Vita to Beta Lehem suggests that the structure was not a church at the time the work was composed. What Alvares meant by the church of Lalibäla remains unclear unless, perhaps, he was referring to Däbrä Sina, making it the “principal” church of the complex and Golgotha “the church of the least buildings here” (Beckingham & Huntingford, Prester John, p. 207). Achille Raffray, who visited Lalibäla in 1873, speaks of ten churches, but includes Däbrä Sina (“Mikael”) with Golgotha and makes no mention of the Sellassie Chapel, of which he was undoubtedly ignorant. Nor does he refer to the church of Bethlehem (Achille Raffray, Les Églises monolithes de la ville de Lalibéla (Abyssinie), Paris, 1882 [hereafter: Raffray, Lalibéla], p. 5).  It seems very likely that the names of some churches have changed over the course of time, and that the use of some of the excavated structures has gone from civil to ecclesiastical. Certainly, as Stuart Munro-Hay has recently pointed out, some of the wooden alters that have been found in Lalibäla “do not … share the present dedications of the churches” (Ethiopia the Unknown Land. A cultural and historical guide, LondonNew York, 2002 [hereafter: Munro-Hay, Unknown Land] , p. 191).  He also questions, in light of King Lalibäla’s Vita, whether some of the attributions current in the fifteenth century were different from those of today (ibid. p. 193), which may have been different again from the time of their original dedication.

[3] Beckingham & Huntingford, Prester John, , vol. 1, p. 226.

[4] Monti della Corte, Lalibelà, p. 15.

[5] Barriviera, chiese in roccia, pp. 5-6, 35-39, 96-100 and pl. 3, 6, 8 – 11bis, 57-8; Monti della Corte, Lalibelà, pp. 54-63 and pl. XVI. There is a possibility that Miguel de Castanhoso entered the Sellassie Chapel in 1543 as he speaks at one point of “a high altar and other altars, all of the same stone” (as quoted in Munro-Hay, Ethiopia, p. 196). The three monolithic “altars” in the chapel (see below) are unique to the known monuments of Lalibäla, with the apparent exception of the monolith in the maqdas of Betä Merqorios (see below, n. 26). Gerhardt Rohlfs  mentions the Sellassie “tomb” in the following context: “Der König Lalibala liegt in der Golgatha-Kirche begraben, wo auch ein anderer berühmter Heiliger Abessiniens, Selasse, seine Grabstätte hat” (Land und Volk in Afrika. Berichte aus den Jahren 1865-1870, Bremen, 1870, p. 143). It is clear from this confusion over the meaning of the term “Selasse” that he never entered the chapel.

[6] Beatrice Playne, St. George for Ethiopia, London, 1954, p. 139. The accusation is unlikely as Bianchi Barriviera was told he could not see into the “tomb of Lalibäla” without “a special authorization from the head of the Ethiopian Church, because it is the tomb of a saint” (Barriviera, chiese in roccia, p. 34, n. 1. See also below, n. 35).  Had force been used to enter the Sellassie Chapel, it could also have been used to see into the tomb, but was not.

[7] Georg Gerster’s classic fish-eye photograph was first published in 1968 (Kirchen im Fels, Stuttgart; English trans. by Richard Hosking, Churches in Rock. Early Christian Art in Ethiopia, London, 1970 [hereafter: Gerster, Churches in Rock], pl. 81).  The photographs accompanying the present article were taken by the author in the company of E. Balicka-Witakowska, S. Chojnacki and P. Henze in 1993. P. Scholz published two photographs of the central altar from the Sellassie Chapel in 1989, but provides no reference to his source (Piotr Scholz, “Bemerkungen zur Ikonologie der sog. ‘Vier apokalyptischen Wesen’ an dem Steinaltar der Dreifaltigkeitskapelle zu Lalibālā”, in Proceedings of the first International Conference on the History of Ethiopian Art. Sponsored by the Royal Asiatic Society. Held at the Warburg Institute of the University of London, October 21 and 22, 1986London: Pindar Press, 1989, pp. 23-29 (hereafter: Scholz, “Vier apokalyptischen Wesen”), figs. 48 & 50.

[8] Gerster, Churches in Rock, p. 100.


[10] As quoted in Jules Leroy, L’Éthiopie. Archéologie et culture, [Desclee de Brouwer], 1973 (hereafter : Leroy, L’Éthiopie), p.137.

[11] Bidder, Lalibela, p. 116.

[12] In Taddese Beyene, ed., Proceedings of the Eighth International Conference of Ethiopian Studies, University of Addis Ababa, 1984, Addis Ababa, 1989, pp. 137-45 (hereafter: Pirenne, “Signification symbolique”).

[13] Gigar Tesfaye, with the collaboration of Jacqueline Pirenne, “ Inscriptions sur bois de trois églises de Lalibala”, Journal of Ethiopian Studies, 17 (1984), pp. 107-43 (hereafter: Gigar, “Inscriptions”). 

[14] Her argument may also have been influenced by that propounded in Scholz, “Vier apokalyptischen Wesen” (above, n. 7).

[15] Munro-Hay, Unknown Land, p. 191. See also below, pp. 25-6.

[16] Emeri van Donzel, “Ethiopia’s Lalibäla and the fall of Jerusalem 1187”, Aethiopica, 1 (1998), pp. 27-49, esp. p. 40 (hereafter: van Donzel, “Ethiopia’s Lalibäla”); Munro-Hay, Unknown Land, pp. 190-94.

[17] Taddesse Tamrat, Church and State in Ethiopia, 1270-1527, Oxford, 1972 (hereafter: Taddesse, Church and State), p. 58. Taddesse’s opinion is perpetuated by Marilyn E. Heldman, “Legends of Lālibalā, The development of an Ethiopian pilgrimage site”, Res, 27 (1995), pp. 25-38 (hereafter: Heldman, “Legends”), (p. 28 &  n. 24).

[18] “L’église rupestre de Berakit”, Annales d’Éthiopie, 9 (1972), pp. 147-88 & pl. XXV-XXVII (hereafter : Lepage, “Berakit”), esp. pp. 167, 179; idem, “Une origine possible des églises d’Éthiopie”, Comptes rendus de l’Académie des Inscriptions et Belles-Lettres, juillet-octobre 1997, Paris, 1998 (hereafter: Lepage, “Origine”), fasc. 3, pp. 199-211 (p. 210).

[19] David Buxton, “The Christian Antiquities of Northern Ethiopia”, Archaeologia, or Miscellaneous Tracts relating to Antiquity, London: Society of Antiquaries, XCII (1947), pp.1-42 (hereafter: Buxton, “Christian Antiquities”).   

[20] Charles Buckingham, “Notes on an unpublished manuscript of Francesco Alvares: Verdadera Informaçam das Terras do Preste Joam das Indias”, Annales d’Ethiopie, 3 (1959), pp.  139-54 (see p. 145). Raffray reports having found a ms. in the Church of Medhane Alem , apparently a Vita of King Lalibäla, in which it is stated that the work was completed in twenty-three years (Raffray, Lalibéla, p. 9).

[21] Paul Henze has reported the recent excavation of seven rock-cut churches: in some cases the work was undertaken by a single individual. Three discovered in 1997 were: Petros & Pawlos (Tsada Amba,Tigray),  Chicheho Gabriel (Wollo-Gondar border), Etissa (Shoa). A group of four found in 1999 in the Saraya region north of Debre Berhan were dedicated to Mikael, Egziabeher Ab, Maryam and Sellassie.

[22] D.R. Buxton, “Ethiopian Medieval Architecture – The Present State of Studies”, Ethiopian Studies, papers read at the Second International Conference of Ethiopian Studies (Manchester University, July 1963), ed. C.F. Beckingham & Edward Ullendorff, Journal of Semitic Studies, 9 (1964), pp. 239-44 (hereafter: Buxton, “Ethiopian Medieval Architecture”), p. 243.

[23] Sergew Hable Sellassie, Ancient and Medieval Ethiopian History to 1270, Addis Ababa, 1972 (hereafter: Sergew H.S., Ethiopian History), p. 278.

[24] See above, n. 2.

[25] Pirenne, “Signification symbolique”, pp. 140-41; Munro-Hay, Unknown Land, p. 191.

[26] Betä Merqorios has a sanctuary containing a monolithic altar or, more likely in view of its height, a mänbärä tabot (measurements taken by a priest in 2002 at the author’s request were: 2.5 x 1.35 x 1.35m), suggesting a construction date of the late 14th or, quite probably, the 15th century (see below, n. 63).  This date, consistent with the reference to the site in the 15th-century Vie de Lalibela as a church “d’une construction différente” (Perruchon, Vie de Lalibela, p. 124), does not preclude the prior existence of a rock-cut structure which might have served a civil role. Monti della Corte speaks of “un luogo di dimora e di rappresentanza” (Lalibelà, p. 27).  Munro-Hay (Unknown Land, p. 194) suggests that it might have been part of the royal residence (see also: Roderick Grierson and Stuart Munro-Hay, The Ark of the Covenant, London, 1999, p. 256).

[27] Betä Giyorgis also has a stepped foundation, but it is not rendered in the Axumite fashion. In a further departure from Axumite forms, this church has no interior transversals marking the corners of doors and windows.

[28] Mario Di Salvo, with texts by Stanislaw Chojnacki & Osvaldo Raineri, Churches of Ethiopia. The Monastery of Nārgā Śellāsē, Milan-New York, 1999, pp. 62-4.

[29] With the exception that there is no NW chamber at Betä Libanos.

[30] Barriviera, chiese in roccia, p. 31.

[31] Lepage, “installations liturgiques”, p. 91 & fig. 8.

[32] Bianchi Barriviera reports that this eastern niche contains a poorly preserved, larger-than-life sized figure, identified by a later inscription as “Maryam” (Barriviera, chiese in roccia, p. 32), but the presence of any such figure is unlikely.

[33] Barriviera, chiese in roccia, p. 33.

[34] Raffray (Lalibéla, p. 7) writes that the head of the recumbent figure lies on a Greek cross, but recent photographs provide little evidence to confirm this point.

[35] Alvares’ description suggests that a tomb monument stood on the surface of the rock floor which could be seen from any of the windows at the west end of the church “at the right of the high altar” (Beckingham & Huntingford, Prester John, vol. 1, p. 207).  What was meant by the “high altar” is unclear in terms of the actual arrangement, wherein the tomb monument stands in the second bay of the north aisle. There is no altar on the raised area represented by the confines of the first bay, although there may once have been one.  Today, the king’s remains are said to lie in a vault under the floor, probably the spot identified by Alvares as “the entrance to the lower chamber”: “In the centre of the body of the church is the mark of a door like a trap door; it is covered up with a large stone, like an altar stone, fitting very closely. They say that this is the entrance to the lower chamber, and that no one goes in, nor does it look as if the stone or door could be lifted. This stone has a hole in the centre which pierces it through; its size is three spans. All the pilgrims [, who come here in infinite numbers for devotion,] put their hands into this stone (where there is hardly room for them), and they say that many miracles are done” (ibid., pp. 207 & 221). Barriviera describes the tomb monument as a “tabot”, below which “close to the northern wall is a rectangular opening measuring 100 x 180cm. dug into the floor and closed on top by a wooden cover”. He was told by the monks that this opening led to the tomb of Lalibäla, to which he was not able to gain access because it would have involved obtaining a special authorization from the head of the Ethiopian Church. As a consequence “we could not check this enclosed space to see whether it differed from others which were open and empty, nor could we clarify Alvares’ text, which does not correspond with what we saw” (Barriviera, chiese in roccia, p. 34 & n. 1; see also Monte della Corte, Lalibelà, p. 61 ).

[36] That this space is surrounded by curtains increases its sanctity, but unlike the sanctuary of an Ethiopian church, non-priests have occasionally obtained access to it.  Golgotha would seem, therefore, not to have a sanctuary per se, although a tabot is kept with the “tomb of Lalibäla” in the north aisle and also with another, commemorating Kidana Mehret (the Pact of Mercy), which stands opposite it in the second bay of the south aisle.

[37] The smallest of these reliefs, its height limited by the existence of a little window above it piercing the wall between Golgotha and Däbrä Sina, is still as large as life. See the illustration in Leroy, L’Éthiopie, p. 139, in which a living priest stands beside the relief.

[38] The inscribed identifications documented by Barriviera are as follows, reading clockwise starting with the figure in the niche in the east wall of the north aisle: 1) Maryam, 2) Yohannes, 3) Qirqos, 4) Giyorgis, 5) Gabra Kristos, 6) Estifanos, and 7) Mikael (Barriviera, chiese in roccia, p. 32).  There is, however, no figure evident in the deep niche designated by Barriviera as no. 7. That the two reliefs seen by Alvares were described as SS Peter and John (Beckingham & Huntingford, Prester John, vol. 1, p. 221), suggest the inscriptions are in fact quite recent as Peter is absent from Barriviera’s list. The four seen by Findlay included SS John, Stephen, George and Kirkos (Louis Findlay, The Monolithic Churches of Lalibela in Ethiopia, Cairo, 1944 [hereafter, Findlay, Monolithic Churches], p.14), that is, those which according to Barriviera’s list correspond to the figures along the south wall and one, Stephen, in the centre of the north wall.  In December 2002, Afe Memhir Alebachu Retta reported in conversation with the author that these inscriptions were added a century ago, that is to say, c. 1900.

[39] Barriviera, chiese in roccia, p. 33 n. 1. Two additional figures would have amounted to a total of nine, which, Barriviera thought, could have represented the well-known Nine Saints to whom are attributed the spreading of the faith in Ethiopia in the fifth century. Munro-Hay finds this “an unlikely theme here”, adding that “if there were nine here, they, with three others in the Sellassie chapel, would represent the twelve apostles” (Munro-Hay, Unknown Land, p. 213).  There are at present only places for two such figures in the Sellassie Chapel, not three, although it has been postulated that a third figure may once have stood in the centre of the chapel’s east wall. There is no sign now of there ever having been a relief sculpture in this central niche. Because the chapel bears the dedication Sellassie (“The Trinity”), those favouring the previous existence of a third figure argue that the three taken together represent the Trinity rather than three of the Twelve Apostles (Gerster, Churches in Rock, p. 102, with reference to fig. 81; Ewald Hein & Brigitte Kleidt, Ethiopia – Christian Africa. Art, Churches and Culture, John M. Deasy, trans., Ratingen: Merlina-Verlag, 1999 [hereafter: Hein & Kleidt, Ethiopia], p. 145).  The fact of the matter is that the single, complete relief figure in the Sellassie Chapel bears the head of an ox and is obviously a representation of one of the four beasts of the Apocalypse, thus negating any association with either the Nine Saints, the Twelve Apostles or the Trinity.

[40] Pirenne, “Signification symbolique”, p. 138.

[41] Barriviera, chiese in roccia, p. 34.

[42] Scholz, “Vier apokalyptischen Wesen”.

[43] Drawings published by both Monti della Corte (Lalibelà, p. 59) and Bianchi Barriviera (chiese in roccia, tav. 10) show a disc containing a relief cross sculpted just below this arch, but none was visible to this writer.

[44] Monti della Corte, Lalibelà, p. 60.  A priest “reluctantly explained” to Bianchi Barriviera that the figure on the left had the head of an ass and that on the right the head of an ox. The sides are obviously transposed and there is absolutely no sign of the head of an ass in the south niche.  Barriviera, who could not have seen what was there, objects to this interpretation (which encouraged Monti della Corte to associate it with a Nativity scene), and supposed instead that the figures had human heads which were destroyed (Barriviera, chiese in roccia, p. 37, n. 1). The occurrence of the standing anthropomorphic figure with the head of an ox is ancient in Christian iconography, as evidenced by the late 8th- 9th-century example occurring on the Soiscél Molaise (National Museum of Ireland, R.4006; see Michelle P. Brown, The Lindisfarne Gospels, Society, Spirituality & the Scribe, London: British Library, 2003, p. 210, fig. 79).

[45] The two windows are very similar in their appearance, although the somewhat more elaborate design of that of the Sellassie chapel suggests that it was inspired by, but carved later than, the Jesus cell window.

[46] The purpose served by this chamber is unknown, although it may possibly have served a function similar to that of the aedicule of the Holy Sepulchre in Jerusalem which covered the rock-cut tomb where Christ’s body was thought to have been laid after the Deposition.

[47] Barriviera, chiese in roccia, pl. 6, no. 3.

[48] This hypothesis is similar to that proposed by Lino Bianchi Barriviera, “Restauri alle chiese di Lalibelà”, Rassegna di Studi Etiopici, 22 (1968 for 1966), pp. 135-46. On p. 138, footnote 2, he writes: “Il ritrovamento, nel corso dei piu recenti lavori di restauro, di tombe scavate nel pavimento di questo corridoio archittonicamente importante, l'esistenza di una croce scolpita nella sua fronte interna est, e altre particolarita gia notate, ci confermano nell'idea che esso possa esser stato originariamente un ingresso, e piu precisamente quello principale, al quale si doveva accedere per una gradinata dalla trincea antistante, il cui piano di calpestio poteva essere a una quota circa corrispondente al ripiano superiore della "Tomba di Adamo"; e che questa e la chiesa di Golgota-Mikael Sellase siano state scavate e scolpite in un tempo successivo, con la conseguente asportazione della gradinata per l'abbassamento della trincea.  Per le considerazioni sopraccenate e per alcune osservazioni relative alle finestre di quest'ultima chiesa, vedi RSE vol XVIII, 1962,  pag. 21, nota 1;  pag.42, nota 1;  pag. 44.  Questa nostra timida ipotesi e confortata dal fatto che anche l'ingresso sud al cortile di Maryam ha una struttura simile, e corrispondenza di quote relative. ”

[49] Amir Harrak, “The Ancient name of Edessa”, Journal of Near Eastern Studies, 51/3 (July, 1992), pp. 209-14.

[50] Ghada Jayyusi-Lehn, “Edessa: A Holy City in the Muslim-Crusader Conflict”, Toronto Studies in Central and Inner Asia, vol. 6 (2004), forthcoming.

[51] Lepage, “Berakit”, p. 150.

[52] On his dating of Degum, see: Lepage, “Berakit”, p. 151; idem, “Découverte d’un art étonnant : les églises éthiopiennes du Xe au XVIe siècle”, Archéologia, 64 (1973), pp. 45-58 (hereafter : Lepage, “églises éthiopiennes”),  (p. 53);  idem, “Le premier art chrétien de l’Éthiopie: Les églises et leur architecture” in “Découverte de  l’Éthiopie chrétienne”, Les Dossiers de l’Archéologie, 8 (1975), pp. 34-59 (hereafter : Lepage, “premier art”), (p. 56).

[53] Lepage, “Berakit”, p. 151 & n. 11.

[54] Lepage, “Berakit”, pp. 152, 155.

[55] Lepage, “installations liturgiques ”, pp. 99-101; Lepage, “églises éthiopiennes”, p. 58.

[56] Mordini gives the measurements of the room as 3.31 x 2.78m, with a height of 2.3m, adding that the monolithic altar stands 1.7m. high by 1m wide (Antonio Mordini, “La chiesa ipogea di Ucrò (Ambà Seneiti) nel Tigrai”, Annali dell’Africa Italiana, II (1939), pp. 519-26 (hereafter: Mordini, “Ucrò”), (p. 523). His fig. 4 on pl. 3, however, indicates the height and width to be similar, probably at 1 m. The monoliths at Enda Maryam Wuqro are so similar in appearance to those at May Kado as to suggest they may have been carved by the same workmen.

[57] Lepage identifies this chamber as a “placard” or cupboard  which could have served as a “sacristie” (Lepage, “installations liturgiques”, p. 99), while in Golgotha the purpose is obviously funerary (Lepage, “premier art”, p. 52; Lepage, “origine”, pp. 207, 210 & n. 27).

[58] The church of Gännätä Maryam, which according to an inscription was decorated with mural paintings c. 1270 under the patronage of Yekuno Amlak, the first king to hold the throne after the fall of the Zaguë dynasty, retains many Axumite characteristics. The new, simpler style would seem, therefore, to have made its appearance at a later period still.

[59] The fourteen include Abba Yohanni (Tembien), Abräha Atsbäha, Bahera Maryam, Däbrä Seyon (Geralta), Maryam Dengelat (Amba Sanayt), Enda Maryam Wuqro (Amba Sanayt), Gabriel Wuqien (Tembien), Iyyesus Wanza (or Gedjet), Iyyesus Weleghesa (Tembien), Johannes Maqudi, May Kado Giyorgis, Mikael Ambo, Mikael Barka, and Tselal Moo. It has not as yet been possible to confirm whether there is a monolithic altar in the central sanctuary of Abräha Atsbäha, Gabriel Wuqien (Tembien), or of Iyyesus Wanza. 

[60] Churches containing two monolithic altars, one in the NE corner and one in the centre of the sanctuary, include Bahera Maryam, Enda Maryam Wuqro, Iyyesus Weleghesa (Tembien), May Kado Giyorgis and Tselal Moo. At Abräha Atsbäha, rather than being a monolith, its equivalent in the NE corner is constructed of stone. Since there is a monolithic altar in the SE corner of Mikael Barka, it is likely that there is another in the NE corner, but its presence has not yet been documented.

[61] Lepage, “installations liturgiques”, pp. 98-101.

[62] Gerster, Churches in Rock, p. 81. In addition to having monolithic altars, the common feature is the superimposed rows of small, blind arcades, which in Däbrä Seyon contain the painted heads of holy figures. Outside the church of Iyyesus Wanza, on the south side in line with its east end, are the remains of what would appear to have been a series of three more monolithic altars which might once have been enclosed in a built structure.

[63] The priests reported to Ewa Balicka-Witakowska in November 2002 that the tomb of SS. Abräha and Atsbäha in the south compartment of the sanctuary is a monolith covered by metal. It must have originated as a monolithic altar and later been designated as the so-called “grave” of the church's patrons. The compartment is closed by a wall with a window-like opening and an entry. It is opened only once a year on the festival day of SS. Abräha and Atsbäha.

[64] The monolithic mänbärä tabot is to be found at Abba Yohanni, Enda Maryam Wuqro, Johannes Magudi, May Kado Giyorgis and Tselal Moo. A cavity to hold the tabot was carved out of the front of the monolithic altar at Bahera Maryam in recent times, in the process destroying the painting of the Virgin Mary between the archangels. Previously, access may only have been from the east side, but the existence of such an opening has not been confirmed. The monolith in the Church of Lalibäla Merqorios, which measures 2.5m high (see above, n. 26), must also belong to this group.

[65] Gerster, Churches in Rock, p. 102; Jean Gire & Roger Schneider, “Étude des églises rupestres du Tigré : Premiers résultats de la mission 1970”, Travaux de la recherche coopérative sur programme R.C.P. 230, Documents pour servir à l’histoire des civilisations éthiopiennes, (series later named Abbay) fasc. 1, 1970, pp. 73-79 and 12 pl. hors texte (pp. 74, 78-79 and pl. 3-4, 11-12); Lepage, “installations liturgiques”, pp. 99-101; Mordini, “Ucrò”, p. 526; Roger Sauter, “Églises rupestres au Tigré”, Annales d’Éthiopie, X (1976), pp. 157-175 (p. 160, no. 8).

[66] René Basset, “Vie d’Abbâ Yohanni”, Bulletin de Correspondance Africaine, III (1884), pp. 433-53 (p. 440).

[67] Monti della Corte, Lalibelà , between pp. 59 & 60; Buxton, “Christian Antiquities”, p. 28/b; Barriviera, chiese in roccia, pl. 9; Gerster, Churches in Rock, p. 102 & fig. 79; Leroy, L’Éthiopie, p. 139; Hein & Kleidt, Ethiopia, p. 146.

[68] See above, n. 38.

[69] Manoel de Almeida, Some Records of Ethiopia 1593-1646, being extracts from “The History of High Ethiopia or Abassia” by Manoel de Almeida together with Bahrey’s “History of the Galla”, ed. C.F. Beckingham & George Wynn Brereton Huntingford, London: Hakluyt Society, Works, second series, vol. CVII, 1954, p. 61.

[70] Lepage, “Berakit”, p. 176; Gerster, Churches in Rock, p. 135 & pl. 191. See also Gigar Tesfaye, “Reconnaissance de Trois Eglises Antérieures à 1314”, Journal of Ethiopian Studies, XII/2 (1974), pp. 57-75 (p. 64).

[71] Sergew H.S., Ethiopian History, p. 275. The author nevertheless concludes that “the art employed on these statues … is Oriental” (ibid.).

[72] Leroy, L’Éthiopie, p. 136.

[73] Leroy, L’Éthiopie, p. 140.

[74] Lepage, “premier art”, p. 46.

[75] Findlay (Monolithic Churches, p. 14) finds them to be “identical in style with similar but smaller ones carved in wood in the Coptic Churches of Cairo”, but suggests no date.

[76] Beckingham & Huntingford, Prester John, vol. 1, p. 221.

[77] J.-B. Coulbeaux indicates that King Lalibäla’s remains lay in the Church of Betä Mariam (Histoire Politique et Religieuse d’Abyssinie Depuis les temps les plus reculés jusqu’à l’avènement de Ménélik II, 2 vols., Paris, 1928 or 1929, vol. 1, p. 269.

[78] The dedication of the Church of Golgotha in Lalibäla to the memory of Christ’s Crucifixion and burial in a rock-cut tomb on the Golgotha hill in Jerusalem makes this site a symbolic representation of the aedicule of the Church of the Holy Sepulchre in Jerusalem. The tradition which places Christ’s Birth in a rock-cut grotto commemorated by the  Church of the Nativity in Bethlehem and his burial in a rock-cut tomb in Jerusalem from which the Resurrection occurred, provides the context for the phenomenon of rock-cut ecclesiastical buildings in the Christian world (see Michael Gervers, “The Iconography of the Cave in Christian and Mithraic Tradition”, in Mysteria Mithrae, ed. Ugo Bianchi, Leiden, 1979, pp. 579-96). The fifteenth-century gadl of King Lalibäla, which is more a work of hagiography than a history, contains many elements which serve to compare the king’s life with that of Christ, while his “burial” beside the “tomb of Christ” in the rock-cut environment of the Church of Golgotha emphasizes the theme of Resurrection attached to the site (see  Michael Gervers, “The Mediterranean Context for the Medieval Rock-Cut Churches of Ethiopia”, Proceedings of the Eighth International Conference of Ethiopian Studies, University of Addis Ababa, 26-30 November 1984, Addis Ababa - Frankfurt am Main, 1988, vol. 1, pp. 171-83; idem, “The Monolithic Church of Wuqro Mäsqäl Krəstos”, Africana Bulletin, 50 (2002), pp. 99-113, esp. pp. 111-13; Lepage, “Origine” pp. 207-10).

[79]van Donzel, “Ethiopia’s Lalibäla”, p. 40.

[80]Heldman, “Legends”, p. 35.

[81]Heldman, “Legends”, p. 33 & fig. 6.

[82] E.A.W. Budge, The Book of Saints (= Synaxary), 4 vols., London, 1928 (reprinted, Hildesheim-New York, 1976), vol. 1, pp. xix-xx; van Donzel, “Ethiopia’s Lalibäla”, pp. 35 & n. 57, 38. Heldman (“Legends”, p. 35) postpones the inclusion until the sixteenth century.

[83] Getatchew Haile and Wm. F. Macomber, A Catalogue of Ethiopian Manuscripts Microfilmed for the Ethiopian Manuscript Microfilm Library, Addis Ababa and for the Hill Monastic Manuscript Library, Collegeville, vol. 5, Collegeville, MN, 1981, pp. 121-22; Heldman, “Legends”, p. 33; Munro-Hay, Unknown Land, p. 190. The example in the British Museum (Or. 719), which may be the oldest surviving copy, is dated to 1434 (C.W. Wright, ed., Catalogue of the Ethiopic manuscripts in the British Museum acquired since the year 1847, London, 1877, nos. 294 & 295. Lepage (“premier art”, p. 45) and van Donzel (“Ethiopia’s Lalibäla”, p. 28, n. 4) attribute King Lalibäla’s gadl to the fourteenth century.

[84] In the second half of the twelfth century, and apparently as late as the fifteenth century when the Gadla Yəmrähannä Krəstos was written, the site was known as Arafah/Adefa (Munro-Hay, Unknown Land, pp. 193-94).  Sergew H.S. thought the change from another name, Werwer, to Roha took place at some point early on in the construction of the site (Ethiopian History, p. 273). On Roha/Warwar, see also Heldman “Legends”, p. 29, and van Donzel, “Ethiopia’s Lalibäla”, pp. 34, 39.  Monti della Corte notes that the first occurrence of Lalibäla as a place name occurs as “Lalabedà” in the mappa mundi published by the Venetian Fra Mauro in 1459 (Lalibelà, p. 21, n. 1).

[85] van Donzel, “Ethiopia’s Lalibäla”, p. 38.

[86] P. Piovanelli, “Les controverses théologiques sous le roi Zar'a Yā'qob  (1434-1468) et la mise en place du monophysisme éthiopien”, in : Alain Le Boulluec, ed.,  La controverse religieuse et ses formes, Paris, 1995, pp. 189-228; Edward Ullendorff, The Ethiopians. An Introduction to Country and People,  2nd ed. London, 1965, p. 105; Getachew Haile, “Religious Controversies and Growth of Ethiopic Literature in the 14th and 15th Centuries”, Oriens Christianus, 65 (1981), pp. 102-36, esp. pp. 131-32; idem, “Ethiopian Saints”, in The Coptic Encyclopedia, vol. 4, NY, 1991, pp. 1050-53; Taddesse, Church and State, pp. 206-47.

[87] Gigar, “Inscriptions”, pp. 114-19.

[88] Gigar, “Inscriptions”, p. 108.

[89] Gianfrancesco Lusini argues that it does, noting that the spiritual father of Ewostatewos (1273-1352) was Dan’el, abbot of Däbrä Maryam on Mount Qorqor, whom he met c. 1280. He further argues that Dan’el, in turn, was the student of ‘Ebna Sanbat’, a name which he interprets to mean “Stone of the Sabbath” or “Son of the Sabbath”.  “This name”, he writes, “leads one to believe it to be a clue to the existence of some controversy about the observance of the ‘Sabbath’” around the middle of the thirteenth century; that is, during the period of Zaguë rule. Lusini refers to the inscription on the tabot published by Gigar Tesfaye as providing further evidence for “a possible ‘Sabbathic’ controversy during the reign of Lālibalā” (Gianfrancesco Lusini, “A new source for the history of Gar’āltā (Ethiopia). The “Life” of Dān’ēl of  Dabra Māryām on Mount Qorqor (KRZ 36)”, Quaderni Utinensi, 8 (15-16) (1996 for 1990), pp. 345-352.  See also: idem, Studi sul Monachesimo Eustaziano (secoli xiv-xv), Naples, 1993, pp. 27-28). If the inscription does date from Lalibäla’s reign, it may be that part of the king’s renown, and consequently his rehabilitation a century or more later, can be attributed to his orthodox position.

[90] Gigar, “Inscriptions”; Munro-Hay, Unknown Land, p. 191. See also p. 5 and n. 15.

[91] Buxton, “Ethiopian Medieval Architecture”, p. 243, n. 1.