Law codes and official lists create an idealized view of the Roman Empire, as it was conceived by lawyers, bureaucrats, and perhaps even emperors. The real world was often different, and sources that refiect local conditions throughout the empire are the historian’s antidote to a surfeit of legal and administrative rhetoric. Some of the best information for what Roman society was like and how the empire worked in practice comes from contemporary letters. The bias towards preserving Christian texts has meant that most surviving correspondence was written by bishops, who frequently took the role of secular civic leaders. From Gaul we have large collections of letters by Sidonius Apollinaris, from Caesarius bishop of Arles in the early sixth century, his contemporary Avitus of Vienne, and Remigius of Reims.64 Other western collections in Latin are those of Ambrose, bishop of Milan from the late fourth century, Augustine in Africa in the early fifth century (including a newly discovered collection of previously unpublished letters), and Ennodius of Pavia, the bishop of Milan, relating to Ostrogothic Italy. In the East, Basil, bishop of Caesarea from 370 to 378, wrote numerous letters to officials at all levels from the praetorian prefect of the East to local censors. The majority of these petitions contained requests for tax exemptions.65 Basil’s closest fifth-century counterpart was Theodoret, bishop of Cyrrhus in north Syria from 423 to c.460. His pleas for Cyrrhus and its citizens included a letter to the Augusta Pulcheria herself (ep. 43), as well as praetorian prefects, various quaestors of the imperial palace, leading patricians of the time ofTheodosius II, and military commanders.66 The major churchmen of the period were usually prolific correspondents. Most of their letters relate to internal church affairs and doctrinal matters, but supply many vital clues to the wider social context in which the church was embedded. Some were selected and edited for wider circulation, such as the main collection of Ambrose’s correspondence, and these were doubtless designed to show the writer in a favorable light.67