Paul the Deacon wrote about two earthquakes that occurred roughly three years apart. The year he provides for the first of these — the Holy Desert Quake — is inconsistent with archaeoseismic evidence from Bet She’an, which establishes a terminus post quem of 749 CE for the event that struck the Jordan Valley and the Sea of Galilee. To avoid chronological confusion, these two earthquakes are referred to by name rather than by date. The Holy Desert Earthquake is described as striking first at 10 a.m. in January, though this time likely corresponds to when the subsequent Talking Mule Quake occurred. The Holy Desert Quake is described as “a great earthquake in Palestine, by the Jordan, and in all of Syria,” during which “an innumerable multitude perished — many tens of thousands,” and “churches and monasteries collapsed.” “The worst was in the wilderness of the Holy City (Jerusalem).”

The Talking Mule Quake is said to have struck Syria in the same year (A.M._a 6241, equivalent to 25 March 748 to 24 March 749 CE) in which Leo the Khazar was born to Constantine V. Paul the Deacon reports that “many died” and that “a spring [moved?].” He adds that “in another place in the mountains, a village moved for about six miles with its walls and homes intact and without any small thing dying.” This “sliding village” appears to be an embellished description of a translational landslide known as a block slide. Paul concludes by writing that “in Mesopotamia, the earth split two thousand [feet?], and out of the chasm came a different soil, which was white and sandy.” “Out of this chasm emerged a spotless mule speaking in a human voice, predicting an invasion by a foreign army into the land of the Arabs — which came true.” Despite the miraculous element of the “Talking Mule” — which became a popular feature of later retellings — the account of an earth fissure and sand boils is seismically credible.