The purpose of this paper is to re-examine two alleged seventh-century dreams of roughly contemporaneous date in order to highlight the manner in which knowledge of their classical antecedents may either hinder or help in the proper interpretation of the same. In the first case, I will discuss the dream allegedly experienced by Maximus the Confessor which had seemed to prophesize the success of the revolt of the exarch Gregory against Constans II in 646. His sharing of this dream was one of the charges against Maximus at his trial in 655. This dream deserves fresh examination because in the words of Neil (2021, p. 170), it appears to be ‘a rare instance in hagio-historiography of a dream that was not veridical, even though it featured angels’. In this case, I argue, a knowledge of classical antecedents may have encouraged both imperial officials and modern commentators to misunderstand the significance of this dream.
In the second case, I will discuss the dream allegedly experienced by the emperor Constans II himself shortly before his defeat by the Arabs at the naval battle of Phoenix in 655. In this case, I will argue that a knowledge of classical antecedent suggests that the key source for this dream, the lost work of Theophilus of Edessa, has completely misunderstood the connection between this dream and what happened next during the battle