Legions

There were from one to three legions based in Egypt in the Principate, when the Roman army empire-wide totalled 25 to 30 legions. Strabo (17.1.12), writing under Augustus, records three in Egypt, but Tacitus (Annals 4.5), writing about the year AD 23, notes only two. By the early third century, when Cassius Dio (55.24) described the deployment of the legions throughout the empire, there was only one, named as II Traiana (see Ritterling 1925a). The legions of the first and early second century can be identified as III Cyrenaica and XXII Deiotariana (see Ritterling 1925b; 1925c).

Strabo describes where the three legions of his day were based, one ‘‘in the City’’ (in the camp at Nikopolis, on the outskirts of Alexandria) and two in the chora (two legions). One of the legions of the chora was at Babylon/Old Cairo and the other probably at Thebes/Luxor (Strabo 17.1.30 for Babylon; Speidel 1982 for Thebes). These locations made strategic sense. Alexandria was a major population center and link between Egypt and the wider Roman Mediterranean. Babylon was a crucial node of land and water transport at the apex of the Delta and Thebes a site for controlling Upper Egypt and securing the southern frontiers. All three were the sites of legionary bases in the later Roman Period, and the surviving archaeological remains date to that later period. Subsequently the remaining two legions were concentrated at Alexandria, which emphasizes the importance of the internal security role of the army.

Funerary epitaphs of individuals, along with a few collective inscriptions, provide evidence for the origins of legionaries. Of particular importance is a fragmentary inscription from Koptos (ILS 2483 = CIL III 6627) recording the names, filiation, and origins of thirty-six legionaries engaged in construction work in the Eastern Desert (Alston 1998: 29-31 provides a useful discussion). Each soldier has the same praenomen (first name) as his father, an unlikely coincidence unless, as Mommsen (the editor) suggested, this was a legal fiction to disguise the fact that these men were not Roman citizens on their recruitment. As a general rule, Roman legions were recruited from Roman citizens, but the eastern legions had often recruited peregrines (non-citizens), a practice dating back to Mark Antony, and reflecting the relative lack of Roman citizens in many eastern provinces, particularly Egypt (Mann 1983, 45). Nearly half of the men in this inscription came from Asia Minor, particularly Galatian cities such as Ankyra (Ankara). Only three soldiers came from the West (Gaul and Italy), reflecting the culturally Greek character of the eastern Roman army. Seven men were recruited in Egypt (six of them in Alexandria) and two more state their origins as castris (‘‘from the camp’’), suggesting that they were sons of soldiers, possibly (but not necessarily) serving in Egypt. Unfortunately the date of the document is disputed, and there are grounds for advancing both Augustan and Flavian dates (Alston, cit. sets out the arguments). Whichever date is accepted, the number of men from Egypt is relatively high and has been taken as an indication that more localized recruitment of legionaries, and increasing recruitment of the sons of soldiers (empire-wide trends) developed earlier in Egypt than elsewhere (Mann 1983: 44-5).

This impression is reinforced by an ad 194 inscription from Alexandria ( CIL III, 6580) that lists the origins of 41 men of II Traiana who would have been recruited some 25 years earlier. They include eight men from Egypt and 24 castris. If we assume that these recruits’ fathers had served in specifically Egyptian legions, this suggests a very high degree of local recruitment. On the other hand, a dedication of AD 157 from Alexandria by 130 men recruited to II Traiana in ad 132-3 (AE 1955: 238 + AE 1969: 633) shows an overwhelming predominance (two-thirds) of men from Africa. Mann (1983: 46-7) argues that this evidence is unrepresentative, reflecting emergency recruitment relating to the Bar-Kochva revolt in Judaea, while

Alston (1998: 44-8) advances an argument that this reflects normal recruitment, and that the importance of Africa as a recruiting ground for the Egyptian legions has been underestimated.