The general question of the nature of social, cultural, and legal relationships between soldiers and civilians is one that has been debated at some length in recent decades (Campbell 1984: 246-63; Alston 1995, 1999; Pollard 1996, 2000). While some scholars have argued for integration, and others for separation, these extremes are more useful as starting points for discussion than as depictions of reality. Undoubtedly both integration and separation coexisted, perhaps in different degrees from province to province and individual to individual, and different kinds of evidence (private letters, official petitions and contracts, upper-class literary commentary on the army) tend to present different perspectives.
Particularly given tendencies to local recruitment, soldiers often lived and worked among the population from which they were drawn, and one might expect to find evidence of continuing social relations between the two groups. Intermarriage might strengthen such relationships. As we have seen, it is likely that there were also economic relationships between soldiers and civilians. On the basis of a wealth of detailed information from papyri from Egyptian sites such as Karanis, Alston (1995, 1999) concluded that such strong social relationships did exist among army veterans
And local civilians, and that district centurions, on detached policing duties in the villages, fitted in well to existing structures of power and patronage, acting as arbitrators in local disputes. There is good evidence from other provinces of veterans integrating into local civilian communities, albeit sometimes with enhanced status and wealth due to their military service (for example, see MacAdam 1983: 113-14 on the villages of southern Syria and Arabia).
On the other hand, soldiers often were used as the lowest level of administration of an empire that did not have many civilian administrators, and they engaged in policing, tax collecting, and low level juridical activity among other things (Davies 1974 still provides a useful survey; see also Isaac 1992: 115-18; Alston 1995: 81-96; Pollard 2000: 85-104). They could legally demand accommodation in the homes of private individuals (hospitium) and requisition animals, vehicles, and labor for official transport duties (angaria) (Isaac 1992: 291-7; Pollard 2000: 104-9). Even the proper conduct of such duties might lead to tensions with the subject population, but there is considerable evidence of abuse of authority, status, and privilege (Campbell 1984: 246-54; Isaac 1992: 269-310; see also Campbell 1994: nos. 286-301 for examples of both duties and abuses). Apuleius’ much-repeated story (Met. 9.39-42 = Campbell 1994: no. 291) of how a legionary robbed and beat a civilian on the pretext of requisitioning his ass is a fictional and humorous one. However, legal scholars (D. 220.127.116.11 [Ulpian] = Campbell 1994: no. 292), Roman officials (an edict ofthe prefect of Egypt, PSI446 = SP 221= Campbell 1994: no. 293), and local communities (the village of Phaena in Syria [IGRR 3.1119 = Campbell 1994: no. 296]) all refer to abuse of hospitium and angaria. Likewise documents from Egypt (e. g. SB 9207 = Campbell 1994: no. 297) record extortion by soldiers, while a writing tablet from Britain (Tab. Vindol. II.344) records an appeal against violent ill-treatment of (probably) a trader by soldiers.
Soldiers also had legal privileges that may have set them apart from civilians. While Juvenal’s complaints on this subject (Sat. 16) undoubtedly were exaggerated for humorous effect, jurists and other legal sources (D. 18.104.22.168 = Campbell 1994: no. 176, for example) envision that a man might join the army to gain an advantage in a forthcoming lawsuit. Likewise veterans had privileges (besides perhaps being relatively wealthy due to their savings andpraemia) that might be resented by some civilians. For example, they were exempt from some taxes (as set out in Domitian’s edict of 94 ce, ILS 9059 = Campbell 1994: no. 341) and some degrading forms of punishment, such as beating with rods and being thrown to the beasts (D. 49.18.1-5 = Campbell 1994: no. 336).
Some scholars (B. D. Shaw 1983; MacMullen 1984c; Pollard 1996; see also papers in Goldsworthy and Haynes 1999) have suggested that the separation of soldiers from civilians was further enhanced by the social exclusivity of the army as a whole and of individual army units. Such exclusivity may have been promoted by common values, links of family, rank, and comradeship as well as shared cultural bonds such as those of religion. At Dura-Europos, for example, soldiers participated together in official unit-level religious ceremonies paying cult to the emperor and a range of other deities, as shown by a papyrus calendar of festivals (the Feriale Duranum) of the 20th cohort of Palmyrenes and a vivid mural of one such ceremony (P. Dura 54 = Fink 1971: no. 117; Cumont 1926: 89-114, pls. xlix-li; Pollard 1996: 221). Soldiers at Dura (and elsewhere) also gathered and worshipped together in private cults such as that of Mithras and Jupiter Dolichenus (Pollard 1996: 221-2), enhancing military solidarity on a private as well as official level. If such an ‘‘inward looking ethos and customary behavior’’ (B. D. Shaw 1983: 151) did dominate the social and cultural relationships of individual soldiers, then it may have made it easier for Roman officials to use the army for oppressive purposes against provincial civilians, and even turn it against a ruling emperor. However, Alston (1999, providing evidence from Egypt) has warned against assuming that the institutional characteristics of the army necessarily superseded individual soldiers’ other social and cultural networks.
Thus there is plenty of evidence for things that might set soldiers and civilians apart. Whether we should take such often anecdotal material as typical is a difficult question. However, Roman soldiers were representatives of an often violent and rapacious imperial state, and we should not assume that the empire was universally popular among its subjects.