The Rock-Hewn Churches of Lalibela

Traditional date

In the highlands of Ethiopia a group of churches is cut whole out of the living rock. Ethiopian tradition ascribes them to the saint-king Lalibela, about 1200; the written sources are late, and archaeologists still debate how long the work took.

At Lalibela, in the mountains of northern Ethiopia, stand churches that were not built but cut out of the rock, with their pillars, windows and doors, as Francisco Álvares described them in the 1520s (Álvares, Prester John of the Indies 53 (pp. 122–131)). The Ethiopian Orthodox Tewahedo Church ascribes them to a king of the same name, Gebre Meskel Lalibela, and venerates him as a saint. Its Synaxarium commemorates him on the twelfth day of the month of Sanê as king of Ethiopia and “seer of the mysteries of heaven”, to whom an angel showed in a dream how to build ten churches together, each different from the others (Ethiopian Synaxarium 12 Sanê (Budge, pp. 995–996)).

A longer telling is his Life, a saint’s life in Ge’ez, of which its first editor knew a copy that he assigned to the fourteenth century (Life of Lalibela Perruchon’s introduction, section IV). It relates that before he was king an angel carried him up through the heavens, where God showed him ten churches, each made of a single stone, and told him that he would be anointed king in order to bring their like out of the earth (Life of Lalibela Perruchon, pp. 86–89). As king, the Life says, he had iron tools forged, asked the workmen to name their own wages and paid them, and bought the ground from its owners. It names the churches: the houses of Mary, of the Cross and of the Virgins, with Debre Sina and Golgotha, cut from one rock; of Gabriel and of Abba Mata; of Mercurius and of Emmanuel; and the house of George, standing alone in the form of a cross. Angels, it says, worked beside the masons by day and alone by night, so that a cubit cut by evening was four by morning (Life of Lalibela Perruchon, pp. 121–125). The Life sets these churches beside Solomon’s temple and tells every Ethiopian to go to the holy city of Roha to see them (Life of Lalibela Perruchon, pp. 126–127).

Álvares, a Portuguese priest, came twice. He names ten churches, gives their measurements, and writes that buildings like them, and so many, are in his opinion not to be found elsewhere in the world. Fearing that no one would believe him, he swears that what he has written is true. He was told that King Lalibela had ordered the work, that it was done in twenty-four years and by “Gibetas, that is, white men”, that the king was held to be a saint, and that his tomb in the church of Golgotha drew many pilgrims (Álvares, Prester John of the Indies 53 (pp. 122–131)).

These are testimonies of a church about its own holy place and of one visitor of the 1520s. They do not settle when each church was cut; that question is taken up in the dating note.

How this is dated

Traditional date. From later tradition or from internal biblical chronology, with no outside anchor.

No document from the time of the cutting survives among the sources used here, and the date on the timeline is the traditional one. It rests on three things.

First, the king. Jules Perruchon, in the introduction to his Vie de Lalibala (1892), reproduces Ignazio Guidi’s reading of the Arabic History of the Patriarchs of Alexandria, according to which a metropolitan ordained in March 1210 found in Ethiopia a king whose name, corrupted in the manuscript, Guidi read as Lalibala (Life of Lalibela Perruchon’s introduction, p. xvi, note). That places his reign in the first years of the thirteenth century.

Second, the tradition that he built the churches. It is told in the Life of Lalibela and in the Synaxarium. Perruchon assigned the oldest copy of the Life known to him to the fourteenth century (Life of Lalibela Perruchon’s introduction, section IV); Nafisa Valieva, “The Portrait of a Queen” (Manuscript of the Month 2016.07, Centre for the Study of Manuscript Cultures, Hamburg), says only that it was composed before the fifteenth. Francisco Álvares heard the same tradition on the spot in the 1520s (Álvares, Prester John of the Indies 53 (pp. 122–131)).

Third, the monuments themselves, which carry no dated inscription cited here and have been read in more than one way. David W. Phillipson, Ancient Churches of Ethiopia: Fourth–Fourteenth Centuries (2009), argued that the oldest rock-cut features at the site go back to the seventh or eighth century and were not made as churches, that the finest churches were added around the eleventh or twelfth, and that the last phase may be dated to King Lalibela’s reign. F.-X. Fauvelle-Aymar and colleagues, “Rock-cut stratigraphy: sequencing the Lalibela churches”, Antiquity 84 (2010), pp. 1135–1150, ask whether the churches belong to one plan or to a centre refashioned over a long time, and set out to read the sequence of construction from the cuts, openings and spoil heaps. The Life counts ten churches; Valieva counts eleven.

Primary sources

Every citation on this page was independently checked against its source on 3 October 2026.

In context

Grows out of

  1. c. AD 330–356, estimated date

Leads to

  1. 1622 to 1632, firm date

Along the threads

Type to search the whole wiki.